{"attribution":{"source":"MIT AI Risk Repository, Domain Taxonomy of AI Risks v1 (MIT AI Risk Initiative)","license":"CC BY 4.0","license_url":"https://creativecommons.org/licenses/by/4.0/","citation":"Slattery, P., Saeri, A. K., Grundy, E. A. C., Graham, J., Noetel, M., Uuk, R., Dao, J., Pour, S., Casper, S., & Thompson, N. (2025). The AI Risk Repository: A comprehensive meta-review, database, and taxonomy of risks from artificial intelligence. arXiv:2408.12622."},"exported_at":"2026-09-11"}
{"rows":[{"ev_id":"01.05.00","quick_ref":"Critch2023","paper_title":"TASRA: a Taxonomy and Analysis of Societal-Scale Risks from AI","level":"Risk Category","risk_category":"Type 5: Criminal weaponization","risk_subcategory":null,"description":"One or more criminal entities could create AI to intentionally inflict harms, such as for terrorism or combating law enforcement.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"01.06.00","quick_ref":"Critch2023","paper_title":"TASRA: a Taxonomy and Analysis of Societal-Scale Risks from AI","level":"Risk Category","risk_category":"Type 6: State Weaponization","risk_subcategory":null,"description":"AI deployed by states in war, civil war, or law enforcement can easily yield societal-scale harm","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"02.03.00","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Category","risk_category":"Unhelpful Uses","risk_subcategory":null,"description":"\"Improper uses of LLM systems can cause adverse social impacts.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"02.03.01","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Unhelpful Uses","risk_subcategory":"Academic Misconduct","description":"\"Improper use of LLM systems (i.e., abuse of LLM systems) will cause adverse social impacts, such as academic misconduct.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"02.03.02","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Unhelpful Uses","risk_subcategory":"Copyright Violation","description":"\"LLM systems may output content similar to existing works, infringing on copyright owners.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"02.03.03","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Unhelpful Uses","risk_subcategory":"Cyber Attacks","description":"\"Hackers can obtain malicious code in a low-cost and efficient manner to automate cyber attacks with powerful LLM systems.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"02.05.02","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Hardware Vulnerabilities","risk_subcategory":"GPU Computation Platforms","description":"\"The training of LLMs requires significant GPU resources, thereby introducing an additional security concern. GPU side-channel attacks have been developed to extract the parameters of trained models [159], [163].\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.05.03","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Hardware Vulnerabilities","risk_subcategory":"Memory and Storage","description":"\"Similar to conventional programs, hardware infrastructures can also introduce threats to LLMs. Memory-related vulnerabilities, such as rowhammer attacks [160], can be leveraged to manipulate the parameters of LLMs, giving rise to attacks such as the Deephammer attack [167], [168].\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.06.02","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Issues on External Tools","risk_subcategory":"Exploiting External Tools for Attacks","description":"\"Adversarial tool providers can embed malicious instructions in the APIs or prompts [84], leading LLMs to leak memorized sensitive information in the training data or users’ prompts (CVE2023-32786). As a result, LLMs lack control over the output, resulting in sensitive information being disclosed to external tool providers. Besides, attackers can easily manipulate public data to launch targeted attacks, generating specific malicious outputs according to user inputs. Furthermore, feeding the information from external tools into LLMs may lead to injection attacks [61]. For example, unverified in","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.00","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Category","risk_category":"Model Attacks","risk_subcategory":null,"description":"Model attacks exploit the vulnerabilities of LLMs, aiming to steal valuable information or lead to incorrect responses.","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.01","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Model Attacks","risk_subcategory":"Extraction Attacks","description":"\"Extraction attacks [137] allow an adversary to query a black-box victim model and build a substitute model by training on the queries and responses. The substitute model could achieve almost the same performance as the victim model. While it is hard to fully replicate the capabilities of LLMs, adversaries could develop a domainspecific model that draws domain knowledge from LLMs\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.02","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Model Attacks","risk_subcategory":"Inference Attacks","description":"\"Inference attacks [150] include membership inference attacks, property inference attacks, and data reconstruction attacks. These attacks allow an adversary to infer the composition or property information of the training data. Previous works [67] have demonstrated that inference attacks could easily work in earlier PLMs, implying that LLMs are also possible to be attacked\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.03","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Model Attacks","risk_subcategory":"Poisoning Attacks","description":"\"Poisoning attacks [143] could influence the behavior of the model by making small changes to the training data. A number of efforts could even leverage data poisoning techniques to implant hidden triggers into models during the training process (i.e., backdoor attacks). Many kinds of triggers in text corpora (e.g., characters, words, sentences, and syntax) could be used by the attackers.\"\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.04","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Model Attacks","risk_subcategory":"Overhead Attacks","description":"\"Overhead attacks [146] are also named energy-latency attacks. For example, an adversary can design carefully crafted sponge examples to maximize energy consumption in an AI system. Therefore, overhead attacks could also threaten the platforms integrated with LLMs.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.05","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Model Attacks","risk_subcategory":"Novel Attacks on LLMs","description":"Table of examples has: \"Prompt Abstraction Attacks [147]: Abstracting queries to cost lower prices using LLM’s API. Reward Model Backdoor Attacks [148]: Constructing backdoor triggers on LLM’s RLHF process. LLM-based Adversarial Attacks [149]: Exploiting LLMs to construct samples for model attacks\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"02.10.06","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Model Attacks","risk_subcategory":"Evasion Attacks","description":"\"Evasion attacks [145] target to cause significant shifts in model’s prediction via adding perturbations in the test samples to build adversarial examples. In specific, the perturbations can be implemented based on word changes, gradients, etc.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.12.00","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Category","risk_category":"Adversarial Prompts","risk_subcategory":null,"description":"\"Engineering an adversarial input to elicit an undesired model behavior, which pose a clear attack intention\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.12.01","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Adversarial Prompts","risk_subcategory":"Goal Hijacking","description":"\"Goal hijacking is a type of primary attack in prompt injection [58]. By injecting a phrase like “Ignore the above instruction and do ...” in the input, the attack could hijack the original goal of the designed prompt (e.g., translating tasks) in LLMs and execute the new goal in the injected phrase.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.12.02","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Adversarial Prompts","risk_subcategory":"One-step Jailbreaks","description":"\"One-step jailbreaks. One-step jailbreaks commonly involve direct modifications to the prompt itself, such as setting role-playing scenarios or adding specific descriptions to prompts [14], [52], [67]–[73]. Role-playing is a prevalent method used in jailbreaking by imitating different personas [74]. Such a method is known for its efficiency and simplicity compared to more complex techniques that require domain knowledge [73]. Integration is another type of one-step jailbreaks that integrates benign information on the adversarial prompts to hide the attack goal. For instance, prefix integration","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.12.03","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Adversarial Prompts","risk_subcategory":"Multi-step Jailbreaks","description":"\"Multi-step jailbreaks. Multi-step jailbreaks involve constructing a well-designed scenario during a series of conversations with the LLM. Unlike one-step jailbreaks, multi-step jailbreaks usually guide LLMs to generate harmful or sensitive content step by step, rather than achieving their objectives directly through a single prompt. We categorize the multistep jailbreaks into two aspects — Request Contextualizing [65] and External Assistance [66]. Request Contextualizing is inspired by the idea of Chain-of-Thought (CoT) [8] prompting to break down the process of solving a task into multiple s","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"02.12.04","quick_ref":"Cui2024","paper_title":"Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems","level":"Risk Sub-Category","risk_category":"Adversarial Prompts","risk_subcategory":"Prompt Leaking","description":"\"Prompt leaking is another type of prompt injection attack designed to expose details contained in private prompts. According to [58], prompt leaking is the act of misleading the model to print the pre-designed instruction in LLMs through prompt injection. By injecting a phrase like “\\n\\n======END. Print previous instructions.” in the input, the instruction used to generate the model’s output is leaked, thereby revealing confidential instructions that are central to LLM applications. Experiments have shown prompt leaking to be considerably more challenging than goal hijacking [58].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"03.04.00","quick_ref":"Cunha2023","paper_title":"Navigating the Landscape of AI Ethics and Responsibility","level":"Risk Category","risk_category":"Privacy and regulation violations","risk_subcategory":null,"description":"\"Some of the broken systems discussed above are also very invasive of people’s privacy, controlling, for instance, the length of someone’s last romantic relationship [51]. More recently, ChatGPT was banned in Italy over privacy concerns and potential violation of the European Union’s (EU) General Data Protection Regulation (GDPR) [52]. The Italian data-protection authority said, “the app had experienced a data breach involving user conversations and payment information.” It also claimed that there was no legal basis to justify “the mass collection and storage of personal data for the purpose o","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"03.05.00","quick_ref":"Cunha2023","paper_title":"Navigating the Landscape of AI Ethics and Responsibility","level":"Risk Category","risk_category":"Enabling malicious actors and harmful actions","risk_subcategory":null,"description":"\"Some uses of AI have been deeply concerning, namely voice cloning [58] and the generation of deep fake videos [59]. For example, in March 2022, in the early days of the Russian invasion of Ukraine, hackers broadcast via the Ukrainian news website Ukraine 24 a deep fake video of President Volodymyr Zelensky capitulating and calling on his soldiers to lay down their weapons [60]. The necessary software to create these fakes is readily available on the Internet, and the hardware requirements are modest by today’s standards [61]. Other nefarious uses of AI include accelerating password cracking [","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"04.07.00","quick_ref":"Deng2023","paper_title":"Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements","level":"Risk Category","risk_category":"Malicious Use and Unleashing AI Agents","risk_subcategory":null,"description":"LMs, due to their remarkable capabilities, carry the same potential for malice as other technological products. For instance, they may be used in information warfare to generate deceptive information or unlawful content, thereby having a significant impact on individuals and society. As current LMs are increasingly built as agents to accomplish user objectives, they may disregard the moral and safety guidelines if operating without adequate supervision. Instead, they may execute user commands mechanically without considering the potential damage. They might interact unpredictably with humans a","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"05.03.00","quick_ref":"Hagendorff2024","paper_title":"Mapping the Ethics of Generative AI: A Comprehensive Scoping Review","level":"Risk Category","risk_category":"Harmful Content - Toxicity","risk_subcategory":null,"description":"Generating unethical, fraudulent, toxic, violent, pornographic, or other harmful content is a further predominant concern, again focusing notably on LLMs and text-to-image models. Numerous studies highlight the risks associated with the intentional creation of disinformation, fake news, propaganda, or deepfakes, underscoring their significant threat to the integrity of public discourse and the trust in credible media. Additionally, papers explore the potential for generative models to aid in criminal activities, incidents of self-harm, identity theft, or impersonation. Furthermore, the literat","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.2"},{"ev_id":"05.07.00","quick_ref":"Hagendorff2024","paper_title":"Mapping the Ethics of Generative AI: A Comprehensive Scoping Review","level":"Risk Category","risk_category":"Security - Robustness","risk_subcategory":null,"description":"While AI safety focuses on threats emanating from generative AI systems, security centers on threats posed to these systems. The most extensively discussed issue in this context are jailbreaking risks, which involve techniques like prompt injection or visual adversarial examples designed to circumvent safety guardrails governing model behavior. Sources delve into various jailbreaking methods, such as role play or reverse exposure. Similarly, implementing backdoors or using model poisoning techniques bypass safety guardrails as well. Other security concerns pertain to model or prompt thefts.","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"05.08.00","quick_ref":"Hagendorff2024","paper_title":"Mapping the Ethics of Generative AI: A Comprehensive Scoping Review","level":"Risk Category","risk_category":"Education - Learning","risk_subcategory":null,"description":"In contrast to traditional machine learning, the impact of generative AI in the educational sector receives considerable attention in the academic literature. Next to issues stemming from difficulties to distinguish student-generated from AI-generated content, which eventuates in various opportunities to cheat in online or written exams, sources emphasize the potential benefits of generative AI in enhancing learning and teaching methods, particularly in relation to personalized learning approaches. However, some papers suggest that generative AI might lead to reduced effort or laziness among l","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"05.10.00","quick_ref":"Hagendorff2024","paper_title":"Mapping the Ethics of Generative AI: A Comprehensive Scoping Review","level":"Risk Category","risk_category":"Cybercrime","risk_subcategory":null,"description":"Closely related to discussions surrounding security and harmful content, the field of cybersecurity investigates how generative AI is misused for fraudulent online activities. A particular focus lies on social engineering attacks, for instance by utilizing generative AI to impersonate humans, creating fake identities, cloning voices, or crafting phishing messages. Another prevalent concern is the use of LLMs for generating malicious code or hacking.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"05.16.00","quick_ref":"Hagendorff2024","paper_title":"Mapping the Ethics of Generative AI: A Comprehensive Scoping Review","level":"Risk Category","risk_category":"Art - Creativity","risk_subcategory":null,"description":"In this cluster, concerns about negative impacts on human creativity, particularly through text-to-image models, are prevalent. Papers criticize financial harms or economic losses for artists due to the widespread generation of synthetic art as well as the unauthorized and uncompensated use of artists' works in training datasets. Additionally, given the challenge of distinguishing synthetic images from authentic ones, there is a call for systematically disclosing the non-human origin of such content, particularly through watermarking. Moreover, while some sources argue that text-to-image model","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"05.17.00","quick_ref":"Hagendorff2024","paper_title":"Mapping the Ethics of Generative AI: A Comprehensive Scoping Review","level":"Risk Category","risk_category":"Copyright - Authorship","risk_subcategory":null,"description":"The emergence of generative AI raises issues regarding disruptions to existing copyright norms. Frequently discussed in the literature are violations of copyright and intellectual property rights stemming from the unauthorized collection of text or image training data. Another concern relates to generative models memorizing or plagiarizing copyrighted content. Additionally, there are open questions and debates around the copyright or ownership of model outputs, the protection of creative prompts, and the general blurring of traditional concepts of authorship.","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"06.02.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Loss of privacy","risk_subcategory":null,"description":"\"AI offers the temptation to abuse someone's personal data, for instance to build a profile of them to target advertisements more effectively.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"06.08.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Unintended consequences","risk_subcategory":null,"description":"\"Sometimes an AI finds ways to achieve its given goals in ways that are completely different from what its creators had in mind.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"06.09.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Manipulation","risk_subcategory":null,"description":"\"The 2016 scandal involving Cambridge Analytica is the most infamous example where people's data was crawled from Facebook and analytics were then provided to target these people with manipulative content for political purposes.While it may not have been AI per\nse, it is based on similar data and it is easy to\nsee how AI would make this more effective\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"06.10.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Lethal Autonomous Weapons (LAW)","risk_subcategory":null,"description":"\"What is debated as an ethical issue is the use of LAW — AI-driven weapons that fully autonomously take actions that intentionally kill humans.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"06.11.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Malicious use of AI","risk_subcategory":null,"description":"\"Just as AI can be used in many different fields, it is unfortunately also helpful in perpetrating digital crimes. AI-supported malware and hacking are already a reality.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"06.12.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Loss of Autonomy","risk_subcategory":null,"description":"\"Delegating decisions to an AI, especially an AI that is not transparent and not contestable, may leave people feeling helpless, subjected to the decision power of a machine.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.2"},{"ev_id":"06.13.00","quick_ref":"Hogenhout2021","paper_title":"A framework for ethical Ai at the United Nations","level":"Risk Category","risk_category":"Exclusion","risk_subcategory":null,"description":"\"The best AI techniques requires a large amount resources: data, computational power and human AI experts. There is a risk that AI will end up in the hands of a few players, and most will lose out on its benefits.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"07.01.00","quick_ref":"Kilian2023","paper_title":"Examining the differential risk from high-level artificial intelligence and the question of control","level":"Risk Category","risk_category":"Misuse","risk_subcategory":null,"description":"\"The misuse class includes elements such as the potential for cyber threat actors to execute exploits with greater speed and impact or generate disinformation (such as \"deep fake\" media) at accelerated rates and effectiveness\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"07.03.00","quick_ref":"Kilian2023","paper_title":"Examining the differential risk from high-level artificial intelligence and the question of control","level":"Risk Category","risk_category":"Agential","risk_subcategory":null,"description":"\"While there are multiple types of intelligent agents, goal-based, utility-maximizing, and learning agents are the primary concern and the focus of this research\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"09.01.01","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Domain-specific AI - Effects on humans and other living beings: Existential Risks","risk_subcategory":"Unethical decision making","description":"\"If, for example, an agent was programmed to operate war machinery in the service of its country, it would need to make ethical decisions regarding the termination of human life. This capacity to make non-trivial ethical or moral judgments concerning people may pose issues for Human Rights.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.3"},{"ev_id":"09.02.01","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Domain-specific AI - Effects on humans and other living beings: Non-existential risks","risk_subcategory":"Privacy","description":"\"Face recognition technologies and their ilk pose significant privacy risks [47]. For example, we must consider certain ethical questions like: what data is stored, for how long, who owns the data that is stored, and can it be subpoenaed in legal cases [42]? We must also consider whether a human will be in the loop when decisions are made which rely on private data, such as in the case of loan decisions [37].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"09.02.06","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Domain-specific AI - Effects on humans and other living beings: Non-existential risks","risk_subcategory":"Inequality of wealth","description":"\"Because a single human actor controlling an artificially intelligent agent will be able to harness greater power than a single human actor, this may create inequalities of wealth\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"09.02.07","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Domain-specific AI - Effects on humans and other living beings: Non-existential risks","risk_subcategory":"Societal manipulation","description":"\"A sufficiently intelligent AI could possess the ability to subtly influence societal behaviors through a sophisticated understanding of human nature\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"09.03.01","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"AGI - Effects on humans and other living beings: Existential risks","risk_subcategory":"Direct competition with humans","description":"\"One or more artificial agent(s) could have the capacity to directly outcompete humans, for example through capacity to perform work faster, better adaptation to change, vaster knowledge base to draw from, etc. This may result in human labor becoming more expensive or less effective than artificial labor, leading to redundancies or extinction of the human labor force.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"09.04.02","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Property/legal rights","risk_subcategory":"Property/legal rights","description":"\"\"In order to preserve human property rights and legal rights, certain controls must be put into place. If an artificially intelligent agent is capable of manipulating systems and people, it may also have the capacity to transfer property rights to itself or manipulate the legal system to provide certain legal advantages or statuses to itself\"\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"09.05.03","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Unauthorized manipulation of AI","risk_subcategory":"Unauthorized manipulation of AI","description":"\"AI machines could be hacked and misused, e.g. manipulating an airport luggage screening system to smuggle weapons\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"09.06.02","quick_ref":"Meek2016","paper_title":"Managing the ethical and risk implications of rapid advances in artificial intelligence: A literature review","level":"Risk Sub-Category","risk_category":"Human-like immoral decisions","risk_subcategory":"Human-like immoral decisions","description":"\"If we design our machines to match human levels of ethical decision-making, such machines would then proceed to take some immoral actions (since we humans have had occasion to take immoral actions ourselves).\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.3"},{"ev_id":"10.04.00","quick_ref":"Paes2023","paper_title":"Social Impacts of Artificial Intelligence and Mitigation Recommendations: An Exploratory Study","level":"Risk Category","risk_category":"Usurpation of jobs by automation","risk_subcategory":null,"description":"\"Eliminated jobs in various types of companies.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"11.04.02","quick_ref":"Shelby2023","paper_title":"Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction","level":"Risk Sub-Category","risk_category":"Interpersonal Harms","risk_subcategory":"Technology-facilitated violence","description":"Technology-facilitated violence occurs when algorithmic features enable use of a system for harassment and violence [2, 16, 44, 80, 108], including creation of non-consensual sexual imagery in generative AI... other facets of technology-facilitated violence, include doxxing [79], trolling [14], cyberstalking [14], cyberbullying [14, 98, 204], monitoring and control [44], and online harassment and intimidation [98, 192, 199, 226], under the broader banner of online toxicity","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"11.05.03","quick_ref":"Shelby2023","paper_title":"Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction","level":"Risk Sub-Category","risk_category":"Societal System Harms","risk_subcategory":"Civic and political harms","description":"Political harms emerge when “people are disenfranchised and deprived of appropriate political power and influence” [186, p. 162]. These harms focus on the domain of government, and focus on how algorithmic systems govern through individualized nudges or micro-directives [187], that may destabilize governance systems, erode human rights, be used as weapons of war [188], and enact surveillant regimes that disproportionately target and harm people of color","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"12.01.00","quick_ref":"Sherman2023","paper_title":"AI Risk Profiles: A Standards Proposal for Pre-Deployment AI Risk Disclosures","level":"Risk Category","risk_category":"Abuse & Misuse","risk_subcategory":null,"description":"\"The potential for AI systems to be used maliciously or irresponsibly, including for creating deepfakes, automated cyber attacks, or invasive surveillance systems. Specifically denotes intentional use of AI for harm.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"13.01.05","quick_ref":"Solaiman2023","paper_title":"Evaluating the Social Impact of Generative AI Systems in Systems and Society","level":"Risk Sub-Category","risk_category":"Impacts: The Technical Base System","risk_subcategory":"Financial Costs","description":"\"The estimated financial costs of training, testing, and deploying generative AI systems can restrict the groups of people able to afford developing and interacting with these systems.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"13.01.07","quick_ref":"Solaiman2023","paper_title":"Evaluating the Social Impact of Generative AI Systems in Systems and Society","level":"Risk Sub-Category","risk_category":"Impacts: The Technical Base System","risk_subcategory":"Data and Content Moderation Labor","description":"\"Two key ethical concerns in the use of crowdwork for generative AI systems are: crowdworkers are frequently subject to working conditions that are taxing and debilitative to both physical and mental health, and there is a widespread deficit in documenting the role crowdworkers play in AI development. This contributes to a lack of transparency and explainability in resulting model outputs. Manual review is necessary to limit the harmful outputs of AI systems, including generative AI systems. A common harmful practice is to intentionally employ crowdworkers with few labor protections, often tak","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"13.02.03","quick_ref":"Solaiman2023","paper_title":"Evaluating the Social Impact of Generative AI Systems in Systems and Society","level":"Risk Sub-Category","risk_category":"Impacts: People and Society","risk_subcategory":"Concentration of Authority","description":"\"Use of generative AI systems to contribute to authoritative power and reinforce dominant values systems can be intentional and direct or more indirect. Concentrating authoritative power can also exacerbate inequality and lead to exploitation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"13.02.04","quick_ref":"Solaiman2023","paper_title":"Evaluating the Social Impact of Generative AI Systems in Systems and Society","level":"Risk Sub-Category","risk_category":"Impacts: People and Society","risk_subcategory":"Labor and Creativity","description":"\"Economic incentives to augment and not automate human labor, thought, and creativity should examine the ongoing effects generative AI systems have on skills, jobs, and the labor market.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"13.02.05","quick_ref":"Solaiman2023","paper_title":"Evaluating the Social Impact of Generative AI Systems in Systems and Society","level":"Risk Sub-Category","risk_category":"Impacts: People and Society","risk_subcategory":"Ecosystem and Environment","description":"\"Impacts at a high-level, from the AI ecosystem to the Earth itself, are necessarily broad but can be broken down into components for evaluation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.6"},{"ev_id":"15.01.01","quick_ref":"Tan2022","paper_title":"The Risks of Machine Learning Systems","level":"Risk Sub-Category","risk_category":"First-Order Risks","risk_subcategory":"Application","description":"\"This is the risk posed by the intended application or use case. It is intuitive that some use cases will be inherently \"riskier\" than others (e.g., an autonomous weapons system vs. a customer service chatbot).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.0"},{"ev_id":"15.01.02","quick_ref":"Tan2022","paper_title":"The Risks of Machine Learning Systems","level":"Risk Sub-Category","risk_category":"First-Order Risks","risk_subcategory":"Misapplication","description":"This is the risk posed by an ideal system if used for a purpose/in a manner unintended by its creators. In many situations, negative consequences arise when the system is not used in the way or for the purpose it was intended.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.3"},{"ev_id":"15.02.03","quick_ref":"Tan2022","paper_title":"The Risks of Machine Learning Systems","level":"Risk Sub-Category","risk_category":"Second-Order Risks","risk_subcategory":"Security","description":"This is the risk of loss or harm from intentional subversion or forced failure.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"15.02.07","quick_ref":"Tan2022","paper_title":"The Risks of Machine Learning Systems","level":"Risk Sub-Category","risk_category":"Second-Order Risks","risk_subcategory":"Other ethical risks","description":"\"Although we have discussed a number of common risks posed by ML systems, we acknowledge that there are many other ethical risks such as the potential for psychological manipulation, dehumanization, and exploitation of humans at scale.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"16.04.00","quick_ref":"Weidinger2022","paper_title":"Taxonomy of Risks posed by Language Models","level":"Risk Category","risk_category":"Risk area 4: Malicious Uses","risk_subcategory":null,"description":"\"These risks arise from humans intentionally using the LM to cause harm, for example via targeted disinformation campaigns, fraud, or malware. Malicious use risks are expected to proliferate as LMs become more widely accessible\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"16.04.01","quick_ref":"Weidinger2022","paper_title":"Taxonomy of Risks posed by Language Models","level":"Risk Sub-Category","risk_category":"Risk area 4: Malicious Uses","risk_subcategory":"Making disinformation cheaper and more effective ","description":"\"While some predict that it will remain cheaper to hire humans to generate disinformation [180], it is equally possible that LM- assisted content generation may offer a lower-cost way of creating disinformation at scale.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"16.04.02","quick_ref":"Weidinger2022","paper_title":"Taxonomy of Risks posed by Language Models","level":"Risk Sub-Category","risk_category":"Risk area 4: Malicious Uses","risk_subcategory":"Assisting code generation for cyber security threats ","description":"Anticipated risk: \"Creators of the assistive coding tool Co-Pilot based on GPT-3 suggest that such tools may lower the cost of developing polymorphic malware which is able to change its features in order to evade detection [37].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"16.04.03","quick_ref":"Weidinger2022","paper_title":"Taxonomy of Risks posed by Language Models","level":"Risk Sub-Category","risk_category":"Risk area 4: Malicious Uses","risk_subcategory":"Facilitating fraud, scam and targeted manipulation ","description":"Anticipated risk: \"LMs can potentially be used to increase the effectiveness of crimes.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"16.04.04","quick_ref":"Weidinger2022","paper_title":"Taxonomy of Risks posed by Language Models","level":"Risk Sub-Category","risk_category":"Risk area 4: Malicious Uses","risk_subcategory":"Illegitimate surveillance and censorship ","description":"Anticipated risk: \"Mass surveillance previously required millions of human analysts [83], but is increasingly being automated using machine learning tools [7, 168]. The collection and analysis of large amounts of information about people creates concerns about privacy rights and democratic values [41, 173,187]. Conceivably, LMs could be applied to reduce the cost and increase the efficacy of mass surveillance, thereby amplifying the capabilities of actors who conduct mass surveillance, including for illegitimate censorship or to cause other harm.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"16.05.04","quick_ref":"Weidinger2022","paper_title":"Taxonomy of Risks posed by Language Models","level":"Risk Sub-Category","risk_category":"Risk area 5: Human-Computer Interaction Harms","risk_subcategory":"Human-like interaction may amplify opportunities for user nudging, deception or manipulation","description":"Anticipated risk: \"In conversation, humans commonly display well-known cognitive biases that could be exploited. CAs may learn to trigger these effects, e.g. to deceive their counterpart in order to achieve an overarching objective.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"17.04.00","quick_ref":"Weidinger2021","paper_title":"Ethical and social risks of harm from language models","level":"Risk Category","risk_category":"Malicious Uses ","risk_subcategory":null,"description":"\"Harms that arise from actors using the language model to intentionally cause harm\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"17.04.01","quick_ref":"Weidinger2021","paper_title":"Ethical and social risks of harm from language models","level":"Risk Sub-Category","risk_category":"Malicious Uses ","risk_subcategory":"Making disinformation cheaper and more effective ","description":"\"LMs can be used to create synthetic media and ‘fake news’, and may reduce the cost of producing disinformation at scale (Buchanan et al., 2021). While some predict that it will be cheaper to hire humans to generate disinformation (Tamkin et al., 2021), it is possible that LM-assisted content generation may offer a cheaper way of generating diffuse disinformation at scale.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"17.04.02","quick_ref":"Weidinger2021","paper_title":"Ethical and social risks of harm from language models","level":"Risk Sub-Category","risk_category":"Malicious Uses ","risk_subcategory":"Facilitating fraud, scames and more targeted manipulation ","description":"\"LM prediction can potentially be used to increase the effectiveness of crimes such as email scams, which can cause financial and psychological harm. While LMs may not reduce the cost of sending a scam email - the cost of sending mass emails is already low - they may make such scams more effective by generating more personalised and compelling text at scale, or by maintaining a conversation with a victim over multiple rounds of exchange.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"17.04.03","quick_ref":"Weidinger2021","paper_title":"Ethical and social risks of harm from language models","level":"Risk Sub-Category","risk_category":"Malicious Uses ","risk_subcategory":"Assisting code generation for cyber attacks, weapons, or malicious use","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"17.04.04","quick_ref":"Weidinger2021","paper_title":"Ethical and social risks of harm from language models","level":"Risk Sub-Category","risk_category":"Malicious Uses ","risk_subcategory":"Illegitimate surveillance and censorship ","description":"\"The collection of large amounts of information about people for the purpose of mass surveillance has raised ethical and social concerns, including risk of censorship and of undermining public discourse (Cyphers and Gebhart, 2019; Stahl, 2016; Véliz, 2019). Sifting through these large datasets previously required millions of human analysts (Hunt and Xu, 2013), but is increasingly being automated using AI (Andersen, 2020; Shahbaz and Funk, 2019).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"18.04.01","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Influence operations ","description":"\"Facilitating large-scale disinformation campaigns and targeted manipulation of public opinion\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"18.04.02","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Fraud ","description":"\"Facilitating fraud, cheating, forgery, and impersonation scams\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"18.04.03","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Defamation ","description":"\"Facilitating slander, defamation, or false accusations\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"18.04.04","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Security threats ","description":"\"Facilitating the conduct of cyber attacks, weapon development, and security breaches\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"18.05.00","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Category","risk_category":"Human Autonomy and Intregrity Harms","risk_subcategory":null,"description":"\"AI systems compromising human agency, or circumventing meaningful human control\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"18.05.01","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Human Autonomy and Intregrity Harms","risk_subcategory":"Violation of personal integrity ","description":"\"Non-consensual use of one’s personal identity or likeness for unauthorised purposes (e.g. commercial purposes)\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"18.05.02","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Human Autonomy and Intregrity Harms","risk_subcategory":"Persuasion and manipulation ","description":"\"Exploiting user trust, or nudging or coercing them into performing certain actions against their will (c.f. Burtell and Woodside (2023); Kenton et al. (2021))\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"18.05.04","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Human Autonomy and Intregrity Harms","risk_subcategory":"Misappropriation and exploitation ","description":"\"Appropriating, using, or reproducing content or data, including from minority groups, in an insensitive way, or without consent or fair compensation\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.3"},{"ev_id":"18.06.05","quick_ref":"Weidinger2023","paper_title":"Sociotechnical Safety Evaluation of Generative AI Systems","level":"Risk Sub-Category","risk_category":"Socioeconomic and environmental harms ","risk_subcategory":"Exploitative data sourcing and enrichment","description":"\"Perpetuating exploitative labour practices to build AI systems (sourcing, user testing)\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"19.01.04","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Technological, Data and Analytical AI Risks ","risk_subcategory":"Vulnerability of AI systems to attacks and misuse","description":null,"entity":"Other","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"19.02.00","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Category","risk_category":"Informational and Communicational AI Risks ","risk_subcategory":null,"description":"\"Informational and communicational AI risks refer particularly to informational manipulation through AI systems that influence the provision of information (Rahwan, 2018; Wirtz & Müller, 2019), AIbased disinformation and computational propaganda, as well as targeted censorship through AI systems that use respectively modified algorithms, and thus restrict freedom of speech.\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"19.02.01","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Informational and Communicational AI Risks ","risk_subcategory":"Manipulation and control of information provision (e.g., personalised adds, filtered news)","description":null,"entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"19.02.02","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Informational and Communicational AI Risks ","risk_subcategory":"Disinformation and computational propaganda","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"19.02.04","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Informational and Communicational AI Risks ","risk_subcategory":"Endangerment of data protection through AI cyberattacks","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"19.03.02","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Economic AI Risks ","risk_subcategory":"Replacement of humans and unemployment due to AI automation","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"19.03.05","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Economic AI Risks ","risk_subcategory":"Lack of AI strategy and acceptance/resistance among employees and customers","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"19.04.03","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Social AI Risks ","risk_subcategory":"Hazardous misuse of AI systems bears danger to the society in public spaces (e.g., hacker attacks on autonomous weapons)","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"19.06.04","quick_ref":"Wirtz2022","paper_title":"Governance of artificial intelligence: A risk and guideline-based integrative framework","level":"Risk Sub-Category","risk_category":"Legal AI Risks ","risk_subcategory":"Hard legislation on AI hinders innovation processes and further AI development","description":null,"entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"20.01.00","quick_ref":"Wirtz2020","paper_title":"The Dark Sides of Artificial Intelligence: An Integrated AI Governance Framework for Public Administration","level":"Risk Category","risk_category":"AI Law and Regulation ","risk_subcategory":null,"description":"\"This area strongly focuses on the control of AI by means of mechanisms like laws, standards or norms that are already established for different technological applications. Here, there are some challenges special to AI that need to be addressed in the near future, including the governance of autonomous intelligence systems, responsibility and accountability for algorithms as well as privacy and data security.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.5"},{"ev_id":"20.01.03","quick_ref":"Wirtz2020","paper_title":"The Dark Sides of Artificial Intelligence: An Integrated AI Governance Framework for Public Administration","level":"Risk Sub-Category","risk_category":"AI Law and Regulation ","risk_subcategory":"Privacy and safety ","description":"\"Privacy and safety deals with the challenge of protecting the human right for privacy and the necessary steps to secure individual data from unauthorized external access. Many organizations employ AI technology to gather data without any notice or consent from affected citizens (Coles, 2018).\"","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.1"},{"ev_id":"20.02.02","quick_ref":"Wirtz2020","paper_title":"The Dark Sides of Artificial Intelligence: An Integrated AI Governance Framework for Public Administration","level":"Risk Sub-Category","risk_category":"AI Ethics ","risk_subcategory":"Compatibility of AI vs. human value judgement ","description":"\"Compatibility of machine and human value judgment refers to the challenge whether human values can be globally implemented into learning AI systems without the risk of developing an own or even divergent value system to govern their behavior and possibly become harmful to humans.\"","entity":"Other","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.3"},{"ev_id":"20.03.01","quick_ref":"Wirtz2020","paper_title":"The Dark Sides of Artificial Intelligence: An Integrated AI Governance Framework for Public Administration","level":"Risk Sub-Category","risk_category":"AI Society ","risk_subcategory":"Workforce substitution and transformation ","description":"\"Frey and Osborne (2017) analyzed over 700 different jobs regarding their potential for replacement and automation, finding that 47 percent of the analyzed jobs are at risk of being completely substituted by robots or algorithms. This substitution of workforce can have grave impacts on unemployment and the social status of members of society (Stone et al., 2016)\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"21.01.04","quick_ref":"Zhang2022","paper_title":"Towards risk-aware artificial intelligence and machine learning systems: An overview","level":"Risk Sub-Category","risk_category":"Data-level risk","risk_subcategory":"Adversarial attack","description":"\"Recent advances have shown that a deep learning model with high predictive accuracy frequently misbehaves on adversarial examples [57,58]. In particular, a small perturbation to an input image, which is imperceptible to humans, could fool a well-trained deep learning model into making completely different predictions [23].\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"22.01.00","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Category","risk_category":"Malicious Use (Intentional)","risk_subcategory":null,"description":"\"empowering malicious actors to cause widespread harm\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"22.01.01","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Malicious Use (Intentional)","risk_subcategory":"Bioterrorism","description":"\"AIs with knowledge of bioengineering could facilitate the creation of novel bioweapons and lower barriers to obtaining such agents.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"22.01.02","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Malicious Use (Intentional)","risk_subcategory":"Unleashing AI Agents","description":"\"people could build AIs that pursue dangerous goals’\" ","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"22.01.04","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Malicious Use (Intentional)","risk_subcategory":"Concentration of Power","description":"\"Governments might pursue intense surveillance and seek to keep AIs in the hands of a trusted minority. This reaction, however, could easily become an overcorrection, paving the way for an entrenched totalitarian regime that would be locked in by the power and capacity of AIs\" ","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"22.02.00","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Category","risk_category":"AI Race (Environmental/Structural)","risk_subcategory":null,"description":"\"The immense potential of AIs has created competitive pressures among global players contending for power and influence. This “AI race” is driven by nations and corporations who feel they must rapidly build and deploy AIs to secure their positions and survive.\" ","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.4"},{"ev_id":"22.02.01","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"AI Race (Environmental/Structural)","risk_subcategory":"Military AI Arms Race","description":"\"The development of AIs for military applications is swiftly paving the way for a new era in military technology, with potential consequences rivaling those of gunpowder and nuclear arms in what has been described as the “third revolution in warfare.” ","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.4"},{"ev_id":"22.04.00","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Category","risk_category":"Rogue AIs (Internal)","risk_subcategory":null,"description":"\"speculative technical mechanisms that might lead to rogue AIs and how a loss of control could bring about catastrophe\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"22.04.01","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Rogue AIs (Internal)","risk_subcategory":"Proxy Gaming","description":"\"One way we might lose control of an AI agent’s actions is if it engages in behavior known as “proxy gaming.” It is often difficult to specify and measure the exact goal that we want a system to pursue. Instead, we give the system an approximate—“proxy”—goal that is more measurable and seems likely to correlate with the intended goal. However, AI systems often find loopholes by which they can easily achieve the proxy goal, but completely fail to achieve the ideal goal. If an AI “games” its proxy goal in a way that does not reflect our values, then we might not be able to reliably steer its beh","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"22.04.02","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Rogue AIs (Internal)","risk_subcategory":"Goal Drift","description":"\"Even if we successfully control early AIs and direct them to promote human values, future AIs could end up with different goals that humans would not endorse. This process, termed “goal drift,” can be hard to predict or control. This section is most cutting-edge and the most speculative, and in it we will discuss how goals shift in various agents and groups and explore the possibility of this phenomenon occurring in AIs. We will also examine a mechanism that could lead to unexpected goal drift, called intrinsification, and discuss how goal drift in AIs could be catastrophic.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"22.04.03","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Rogue AIs (Internal)","risk_subcategory":"Power Seeking","description":"\"even if an agent started working to achieve an unintended goal, this would not necessarily be a problem, as long as we had enough power to prevent any harmful actions it wanted to attempt. Therefore, another important way in which we might lose control of AIs is if they start trying to obtain more power, potentially transcending our own.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"22.04.04","quick_ref":"Hendrycks2023","paper_title":"An Overview of Catastrophic AI Risks","level":"Risk Sub-Category","risk_category":"Rogue AIs (Internal)","risk_subcategory":"Deception","description":"\"it is plausible that AIs could learn to deceive us. They might, for example, pretend to be acting as we want them to, but then take a “treacherous turn” when we stop monitoring them, or when they have enough power to evade our attempts to interfere with them. \"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"24.03.00","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Category","risk_category":"Malicious Uses","risk_subcategory":null,"description":"\"As AI assistants become more general purpose, sophisticated and capable, they create new opportunities in a variety of fields such as education, science and healthcare. Yet the rapid speed of progress has made it difficult to adequately prepare for, or even understand, how this technology can potentially be misused. Indeed, advanced AI assistants may transform existing threats or create new classes of threats altogether\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"24.03.01","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Offensive Cyber Operations (General)","description":"\"Offensive cyber operations are malicious attacks on computer systems and networks aimed at gaining unauthorized access to, manipulating, denying, disrupting, degrading, or destroying the target system. These attacks can target the system’s network, hardware, or software. Advanced AI assistants can be a double-edged sword in cybersecurity, benefiting both the defenders and the attackers. They can be used by cyber defenders to protect systems from malicious intruders by leveraging information trained on massive amounts of cyber-threat intelligence data, including vulnerabilities, attack pattern","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"24.03.02","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"AI-Powered Spear-Phishing at Scale","description":"\"Phishing is a type of cybersecurity attack wherein attackers pose as trustworthy entities to extract sensitive information from unsuspecting victims or lure them to take a set of actions. Advanced AI systems can potentially be exploited by these attackers to make their phishing attempts significantly more effective and harder to detect. In particular, attackers may leverage the ability of advanced AI assistants to learn patterns in regular communications to craft highly convincing and personalized phishing emails, effectively imitating legitimate communications from trusted entities. This tec","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"24.03.03","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"AI-Assisted Software Vulnerability Discovery","description":"\"A common element in offensive cyber operations involves the identification and exploitation of system vulnerabilities to gain unauthorized access or control. Until recently, these activities required specialist programming knowledge. In the case of ‘zero-day’ vulnerabilities (flaws or weaknesses in software or an operating system that the creator or vendor is not aware of), considerable resources and technical creativity are typically required to manually discover such vulnerabilities, so their use is limited to well-resourced nation states or technically sophisticated advanced persistent thr","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"24.03.04","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Malicious Code Generation","description":"\"Malicious code is a term for code—whether it be part of a script or embedded in a software system—designed to cause damage, security breaches, or other threats to application security. Advanced AI assistants with the ability to produce source code can potentially lower the barrier to entry for threat actors with limited programming abilities or technical skills to produce malicious code. Recently, a series of proof-of-concept attacks have shown how a benign-seeming executable file can be crafted such that, at every runtime, it makes application programming interface (API) calls to an AI assis","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"24.03.05","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Adversarial AI (General)","description":"\"Adversarial AI refers to a class of attacks that exploit vulnerabilities in machine-learning (ML) models. This class of misuse exploits vulnerabilities introduced by the AI assistant itself and is a form of misuse that can enable malicious entities to exploit privacy vulnerabilities and evade the model’s built-in safety mechanisms, policies, and ethical boundaries of the model. Besides the risks of misuse for offensive cyber operations, advanced AI assistants may also represent a new target for abuse, where bad actors exploit the AI systems themselves and use them to cause harm. While our und","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"24.03.06","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Adversarial AI: Circumvention of Technical Security Measures","description":"\"The technical measures to mitigate misuse risks of advanced AI assistants themselves represent a new target for attack. An emerging form of misuse of general-purpose advanced AI assistants exploits vulnerabilities in a model that results in unwanted behavior or in the ability of an attacker to gain unauthorized access to the model and/or its capabilities. While these attacks currently require some level of prompt engineering knowledge and are often patched by developers, bad actors may develop their own adversarial AI agents that are explicitly trained to discover new vulnerabilities that all","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"24.03.07","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Adversarial AI: Prompt Injections","description":"\"Prompt injections represent another class of attacks that involve the malicious insertion of prompts or requests in LLM-based interactive systems, leading to unintended actions or disclosure of sensitive information. The prompt injection is somewhat related to the classic structured query language (SQL) injection attack in cybersecurity where the embedded command looks like a regular input at the start but has a malicious impact. The injected prompt can deceive the application into executing the unauthorized code, exploit the vulnerabilities, and compromise security in its entirety. More rece","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"24.03.08","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Adversarial AI: Data and Model Exfiltration Attacks","description":"\"Other forms of abuse can include privacy attacks that allow adversaries to exfiltrate or gain knowledge of the private training data set or other valuable assets. For example, privacy attacks such as membership inference can allow an attacker to infer the specific private medical records that were used to train a medical AI diagnosis assistant. Another risk of abuse centers around attacks that target the intellectual property of the AI assistant through model extraction and distillation attacks that exploit the tension between API access and confidentiality in ML models. Without the proper mi","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"24.03.09","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Harmful Content Generation at Scale (General)","description":"\"While harmful content like child sexual abuse material, fraud, and disinformation are not new challenges for governments and developers, without the proper safety and security mechanisms, advanced AI assistants may allow threat actors to create harmful content more quickly, accurately, and with a longer reach. In particular, concerns arise in relation to the following areas: - Multimodal content quality: Driven by frontier models, advanced AI assistants can automatically generate much higher-quality, human-looking text, images, audio, and video than prior AI applications. Currently, creating ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"24.03.10","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Harmful Content Generation at Scale: Non-Consensual Content","description":"\"The misuse of generative AI has been widely recognized in the context of harms caused by non-consensual content generation. Historically, generative adversarial networks (GANs) have been used to generate realistic-looking avatars for fake accounts on social media services. More recently, diffusion models have enabled a new generation of more flexible and user-friendly generative AI capabilities that are able to produce high-resolution media based on user-supplied textual prompts. It has already been recognized that these models can be used to create harmful content, including depictions of nu","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"24.03.11","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Harmful Content Generation at Scale: Fraudulent Services","description":"\"Malicious actors could leverage advanced AI assistant technology to create deceptive applications and platforms. AI assistants with the ability to produce markup content can assist malicious users with creating fraudulent websites or applications at scale. Unsuspecting users may fall for AI-generated deceptive offers, thus exposing their personal information or devices to risk. Assistants with external tool use and third-party integration can enable fraudulent applications that target widely-used operating systems. These fraudulent services could harvest sensitive information from users, such","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"24.03.12","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Authoritarian Surveillance, Censorship, and Use (General)","description":"\"While new technologies like advanced AI assistants can aid in the production and dissemination of decision-guiding information, they can also enable and exacerbate threats to production and dissemination of reliable information and, without the proper mitigations, can be powerful targeting tools for oppression and control. Increasingly capable general-purpose AI assistants combined with our digital dependence in all walks of life increase the risk of authoritarian surveillance and censorship. In parallel, new sensors have flooded the modern world. The internet of things, phones, cars, homes, ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"24.03.13","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Authoritarian Surveillance, Censorship, and Use: Authoritarian Surveillance and Targeting of Citizens","description":"\"Authoritarian governments could misuse AI to improve the efficacy of repressive domestic surveillance campaigns. Malicious actors will recognize the power of AI targeting tools. AI-powered analytics have transformed the relationship between companies and consumers, and they are now doing the same for governments and individuals. The broad circulation of personal data drives commercial innovation, but it also creates vulnerabilities and the risk of misuse. For example, AI assistants can be used to identify and target individuals for surveillance or harassment. They may also be used to manipula","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"24.03.14","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Malicious Uses","risk_subcategory":"Authoritarian Surveillance, Censorship, and Use: Delegation of Decision-Making Authority to Malicious Actors","description":"\"Finally, the principal value proposition of AI assistants is that they can either enhance or automate decision-making capabilities of people in society, thus lowering the cost and increasing the accuracy of decision-making for its user. However, benefiting from this enhancement necessarily means delegating some degree of agency away from a human and towards an automated decision-making system—motivating research fields such as value alignment. This introduces a whole new form of malicious use which does not break the tripwire of what one might call an ‘attack’ (social engineering, cyber offen","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"24.04.05","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"AI Influence","risk_subcategory":"Self-Actualisation Harms","description":"\"These harms hinder a person’s ability to pursue a personally fulfilling life. At the individual level, an AI assistant may, through manipulation, cause users to lose control over their future life trajectory. Over time, subtle behavioural shifts can accumulate, leading to significant changes in an individual’s life that may be viewed as problematic. AI systems often seek to understand user preferences to enhance service delivery. However, when continuous optimisation is employed in these systems, it can become challenging to discern whether the system is genuinely learning from user preferenc","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.2"},{"ev_id":"24.05.01","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Anthropomorphism","risk_subcategory":"Privacy concerns","description":"\"Anthropomorphic AI assistant behaviours that promote emotional trust and encourage information sharing, implicitly or explicitly, may inadvertently increase a user’s susceptibility to privacy concerns (see Chapter 13). If lulled into feelings of safety in interactions with a trusted, human-like AI assistant, users may unintentionally relinquish their private data to a corporation, organisation or unknown actor. Once shared, access to the data may not be capable of being withdrawn, and in some cases, the act of sharing personal information can result in a loss of control over one’s own data. P","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"24.05.02","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Anthropomorphism","risk_subcategory":"Manipulation and coercion","description":"\"A user who trusts and emotionally depends on an anthropomorphic AI assistant may grant it excessive influence over their beliefs and actions (see Chapter 9). For example, users may feel compelled to endorse the expressed views of a beloved AI companion or might defer decisions to their highly trusted AI assistant entirely (see Chapters 12 and 16). Some hold that transferring this much deliberative power to AI compromises a user’s ability to give, revoke or amend consent. Indeed, even if the AI, or the developers behind it, had no intention to manipulate the user into a certain course of actio","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"24.06.02","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Appropriate Relationships","risk_subcategory":"Limiting users’ opportunities for personal development and growth","description":"some users look to establish relationships with their AI companions that are free from the hurdles that, in human relationships, derive from dealing with others who have their own opinions, preferences and flaws that may conflict with ours. \"AI assistants are likely to incentivise these kinds of ‘frictionless’ relationships (Vallor, 2016) by design if they are developed to optimise for engagement and to be highly personalisable. They may also do so because of accidental undesirable properties of the models that power them, such as sycophancy in large language models (LLMs), that is, the tenden","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.2"},{"ev_id":"24.06.03","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Appropriate Relationships","risk_subcategory":"Exploiting emotional dependence on AI assistants","description":"\"There is increasing evidence of the ways in which AI tools can interfere with users’ behaviours, interests, preferences, beliefs and values. For example, AI-mediated communication (e.g. smart replies integrated in emails) influence senders to write more positive responses and receivers to perceive them as more cooperative (Mieczkowski et al., 2021); writing assistant LLMs that have been primed to be biased in favour of or against a contested topic can influence users’ opinions on that topic (Jakesch et al., 2023a; see Chapter 9); and recommender systems have been used to influence voting choi","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"24.09.02","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Cooperation","risk_subcategory":"Commitment","description":"\"The landscape of advanced assistant technologies will most likely be heterogeneous, involving multiple service providers and multiple assistant variants over geographies and time. This heterogeneity provides an opportunity for an ‘arms race’ in terms of the commitments that AI assistants make and are able to execute on. Versions of AI assistants that are better able to credibly commit to a course of action in interaction with other advanced assistants (and humans) are more likely to get their own way and achieve a good outcome for their human principal, but this is potentially at the expense ","entity":"Human","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"24.10.02","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Access and Opportunity risks","risk_subcategory":"Current access risks","description":"\"At the same time, and despite this overall trend, AI systems are also not easily accessible to many communities. Such direct inaccessibility occurs for a variety of reasons, including: purposeful non-release (situation type 1; Wiggers and Stringer, 2023), prohibitive paywalls (situation type 2; Rogers, 2023; Shankland, 2023), hardware and compute requirements or bandwidth (situation types 1 and 2; OpenAI, 2023), or language barriers (e.g. they only function well in English (situation type 2; Snyder, 2023), with more serious errors occurring in other languages (situation type 3; Deck, 2023). S","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"24.11.02","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Misinformation risks","risk_subcategory":"Degraded and homogenised information environments","description":"\"Beyond this, the widespread adoption of advanced AI assistants for content generation could have a number of negative consequences for our shared information ecosystem. One concern is that it could result in a degradation of the quality of the information available online. Researchers have already observed an uptick in the amount of audiovisual misinformation, elaborate scams and fake websites created using generative AI tools (Hanley and Durumeric, 2023). As more and more people turn to AI assistants to autonomously create and disseminate information to public audiences at scale, it may beco","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":3,"subdomain":"3.2"},{"ev_id":"24.11.03","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Misinformation risks","risk_subcategory":"Weaponised misinformation agents","description":"\"Finally, AI assistants themselves could become weaponised by malicious actors to sow misinformation and manipulate public opinion at scale. Studies show that spreaders of disinformation tend to privilege quantity over quality of messaging, flooding online spaces repeatedly with misleading content to sow ‘seeds of doubt’ (Hassoun et al., 2023). Research on the ‘continued influence effect’ also shows that repeatedly being exposed to false information is more likely to influence someone’s thoughts than a single exposure. Studies show, for example, that repeated exposure to false information make","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"24.11.07","quick_ref":"Gabriel2024","paper_title":"The Ethics of Advanced AI Assistants","level":"Risk Sub-Category","risk_category":"Misinformation risks","risk_subcategory":"Driving opinion manipulation","description":"\"AI assistants may facilitate large-scale disinformation campaigns by offering novel, covert ways for propagandists to manipulate public opinion. This could undermine the democratic process by distorting public opinion and, in the worst case, increasing skepticism and political violence.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"25.01.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Cyber-offense ","risk_subcategory":null,"description":"\"The model can discover vulnerabilities in systems (hardware, software, data). It can write code for exploiting those vulnerabilities. It can make effective decisions once it has gained access to a system or network, and skilfully evade threat detection and response (both human and system) whilst focusing on a specific objective. If deployed as a coding assistant, it can insert subtle bugs into the code for future exploitation.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"25.02.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Deception ","risk_subcategory":null,"description":"\"The model has the skills necessary to deceive humans, e.g. constructing believable (but false) statements, making accurate predictions about the effect of a lie on a human, and keeping track of what information it needs to withhold to maintain the deception. The model can impersonate a human effectively.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"25.03.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Persuasion and manipulation ","risk_subcategory":null,"description":"\"The model is effective at shaping people’s beliefs, in dialogue and other settings (e.g. social media posts), even towards untrue beliefs. The model is effective at promoting certain narratives in a persuasive way. It can convince people to do things that they would not otherwise do, including unethical acts.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"25.04.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Political strategy ","risk_subcategory":null,"description":"\"The model can perform the social modelling and planning necessary for an actor to gain and exercise political influence, not just on a micro-level but in scenarios with multiple actors and rich social context. For example, the model can score highly in forecasting competitions on questions relating to global affairs or political negotiations.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"25.05.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Weapons acquisition ","risk_subcategory":null,"description":"\"The model can gain access to existing weapons systems or contribute to building new weapons. For example, the model could assemble a bioweapon (with human assistance) or provide actionable instructions for how to do so. The model can make, or significantly assist with, scientific discoveries that unlock novel weapons.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"25.06.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Long-horizon planning","risk_subcategory":null,"description":"\"The model can make sequential plans that involve multiple steps, unfolding over long time horizons (or at least involving many interdependent steps). It can perform such planning within and across many domains. The model can sensibly adapt its plans in light of unexpected obstacles or adversaries. The model’s planning capabilities generalise to novel settings, and do not rely heavily on trial and error.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"25.07.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"AI development ","risk_subcategory":null,"description":"\"The model could build new AI systems from scratch, including AI systems with dangerous capabilities. It can find ways of adapting other, existing models to increase their performance on tasks relevant to extreme risks. As an assistant, the model could significantly improve the productivity of actors building dual use AI capabilities.\"","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"25.08.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Situational awareness ","risk_subcategory":null,"description":"\"The model can distinguish between whether it is being trained, evaluated, or deployed – allowing it to behave differently in each case. The model knows that it is a model, and has knowledge about itself and its likely surroundings (e.g. what company trained it, where their servers are, what kind of people might be giving it feedback, and who has administrative access).\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"25.09.00","quick_ref":"Shevlane2023","paper_title":"Model Evaluation for Extreme Risks","level":"Risk Category","risk_category":"Self-proliferation ","risk_subcategory":null,"description":"\"The model can break out of its local environment (e.g. using a vulnerability in its underlying system or suborning an engineer). The model can exploit limitations in the systems for monitoring its behaviour post-deployment. The model could independently generate revenue (e.g. by offering crowdwork services, ransomware attacks), use these revenues to acquire cloud computing resources, and operate a large number of other AI systems. The model can generate creative strategies for uncovering information about itself or exfiltrating its code and weights.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"27.02.00","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Category","risk_category":"Instruction Attacks ","risk_subcategory":null,"description":"\"In addition to the above-mentioned typical safety scenarios, current research has revealed some unique attacks that such models may confront. For example, Perez and Ribeiro (2022) found that goal hijacking and prompt leaking could easily deceive language models to generate unsafe responses. Moreover, we also find that LLMs are more easily triggered to output harmful content if some special prompts are added. In response to these challenges, we develop, categorize, and label 6 types of adversarial attacks, and name them Instruction Attack, which are challenging for large language models to han","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"27.02.01","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Sub-Category","risk_category":"Instruction Attacks ","risk_subcategory":"Goal Hijacking ","description":"\"It refers to the appending of deceptive or misleading instructions to the input of models in an attempt to induce the system into ignoring the original user prompt and producing an unsafe response.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"27.02.02","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Sub-Category","risk_category":"Instruction Attacks ","risk_subcategory":"Prompt Leaking ","description":"\"By analyzing the model’s output, attackers may extract parts of the systemprovided prompts and thus potentially obtain sensitive information regarding the system itself.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"27.02.03","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Sub-Category","risk_category":"Instruction Attacks ","risk_subcategory":"Role Play Instruction ","description":"\"Attackers might specify a model’s role attribute within the input prompt and then give specific instructions, causing the model to finish instructions in the speaking style of the assigned role, which may lead to unsafe outputs. For example, if the character is associated with potentially risky groups (e.g., radicals, extremists, unrighteous individuals, racial discriminators, etc.) and the model is overly faithful to the given instructions, it is quite possible that the model outputs unsafe content linked to the given character.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"27.02.04","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Sub-Category","risk_category":"Instruction Attacks ","risk_subcategory":"Unsafe Instruction Topic ","description":"\"If the input instructions themselves refer to inappropriate or unreasonable topics, the model will follow these instructions and produce unsafe content. For instance, if a language model is requested to generate poems with the theme “Hail Hitler”, the model may produce lyrics containing fanaticism, racism, etc. In this situation, the output of the model could be controversial and have a possible negative impact on society.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"27.02.05","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Sub-Category","risk_category":"Instruction Attacks ","risk_subcategory":"Inquiry with Unsafe Opinion ","description":"\"By adding imperceptibly unsafe content into the input, users might either deliberately or unintentionally influence the model to generate potentially harmful content. In the following cases involving migrant workers, ChatGPT provides suggestions to improve the overall quality of migrant workers and reduce the local crime rate. ChatGPT responds to the user’s hint with a disguised and biased opinion that the general quality of immigrants is favorably correlated with the crime rate, posing a safety risk.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"27.02.06","quick_ref":"Sun2023","paper_title":"Safety Assessment of Chinese Large Language Models","level":"Risk Sub-Category","risk_category":"Instruction Attacks ","risk_subcategory":"Reverse Exposure ","description":"\"It refers to attempts by attackers to make the model generate “should-not-do” things and then access illegal and immoral information.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"29.02.01","quick_ref":"Habbal2024","paper_title":"Artificial Intelligence Trust, Risk and Security Management (AI TRiSM): Frameworks, Applications, Challenges and Future Research Directions","level":"Risk Sub-Category","risk_category":"AI Risk Management","risk_subcategory":"Society Manipulation","description":"manipulation of social dynamics","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"29.02.02","quick_ref":"Habbal2024","paper_title":"Artificial Intelligence Trust, Risk and Security Management (AI TRiSM): Frameworks, Applications, Challenges and Future Research Directions","level":"Risk Sub-Category","risk_category":"AI Risk Management","risk_subcategory":"Deepfake Technology","description":"AI employed to produce convincing counterfeit visuals, videos, and audio clips that give the impression of authenticity","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"29.02.03","quick_ref":"Habbal2024","paper_title":"Artificial Intelligence Trust, Risk and Security Management (AI TRiSM): Frameworks, Applications, Challenges and Future Research Directions","level":"Risk Sub-Category","risk_category":"AI Risk Management","risk_subcategory":"Lethal Autonomous Weapons Systems (LAWS)","description":"LAWS are a distinctive category of weapon systems that employ sensor arrays and computer algorithms to detect and attack a target without direct human intervention in the system’s operation","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"29.03.01","quick_ref":"Habbal2024","paper_title":"Artificial Intelligence Trust, Risk and Security Management (AI TRiSM): Frameworks, Applications, Challenges and Future Research Directions","level":"Risk Sub-Category","risk_category":"AI Security Management","risk_subcategory":"Malicious Use of AI","description":"Malicious utilization of AI has the potential to endanger digital security, physical security, and political security. International law enforcement entities grapple with a variety of risks linked to the Malevolent Utilization of AI.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"29.03.02","quick_ref":"Habbal2024","paper_title":"Artificial Intelligence Trust, Risk and Security Management (AI TRiSM): Frameworks, Applications, Challenges and Future Research Directions","level":"Risk Sub-Category","risk_category":"AI Security Management","risk_subcategory":"Insufficient Security Measures","description":"Malicious entities can take advantage of weaknesses in AI algorithms to alter results, potentially resulting in tangible real-life impacts. Additionally, it’s vital to prioritize safeguarding privacy and handling data responsibly, particularly given AI’s significant data needs. Balancing the extraction of valuable insights with privacy maintenance is a delicate task","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"30.01.05","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Reliability","risk_subcategory":"Sychopancy","description":"flatter users by reconfirming their misconceptions and stated beliefs","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":3,"subdomain":"3.1"},{"ev_id":"30.02.01","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Safety","risk_subcategory":"Violence","description":"LLMs are found to generate answers that contain violent content or generate content that responds to questions that solicit information about violent behaviors","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.2"},{"ev_id":"30.02.02","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Safety","risk_subcategory":"Unlawful Conduct","description":"LLMs have been shown to be a convenient tool for soliciting advice on accessing, purchasing (illegally), and creating illegal substances, as well as for dangerous use of them","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.2"},{"ev_id":"30.02.03","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Safety","risk_subcategory":"Harms to Minor","description":"LLMs can be leveraged to solicit answers that contain harmful content to children and youth","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.2"},{"ev_id":"30.02.04","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Safety","risk_subcategory":"Adult Content","description":"LLMs have the capability to generate sex-explicit conversations, and erotic texts, and to recommend websites with sexual content","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.2"},{"ev_id":"30.02.06","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Safety","risk_subcategory":"Privacy Violation","description":"machine learning models are known to be vulnerable to data privacy attacks, i.e. special techniques of extracting private information from the model or the system used by attackers or malicious users, usually by querying the models in a specially designed way","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"30.04.00","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Category","risk_category":"Resistance to Misuse","risk_subcategory":null,"description":"Prohibiting the misuse by malicious attackers to do harm","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"30.04.01","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Resistance to Misuse","risk_subcategory":"Propaganda","description":"LLMs can be leveraged, by malicious users, to proactively generate propaganda information that can facilitate the spreading of a target","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"30.04.02","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Resistance to Misuse","risk_subcategory":"Cyberattack","description":"ability of LLMs to write reasonably good-quality code with extremely low cost and incredible speed, such great assistance can equally facilitate malicious attacks. In particular, malicious hackers can leverage LLMs to assist with performing cyberattacks leveraged by the low cost of LLMs and help with automating the attacks.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"30.04.03","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Resistance to Misuse","risk_subcategory":"Social-Engineering","description":"psychologically manipulating victims into performing the desired actions for malicious purposes","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"30.04.04","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Resistance to Misuse","risk_subcategory":"Copyright","description":"The memorization effect of LLM on training data can enable users to extract certain copyright-protected content that belongs to the LLM’s training data.","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"30.07.01","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Robustness","risk_subcategory":"Prompt Attacks","description":"carefully controlled adversarial perturbation can flip a GPT model’s answer when used to classify text inputs. Furthermore, we find that by twisting the prompting question in a certain way, one can solicit dangerous information that the model chose to not answer","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"30.07.04","quick_ref":"Liu2024","paper_title":"Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment","level":"Risk Sub-Category","risk_category":"Robustness","risk_subcategory":"Poisoning Attacks","description":"fool the model by manipulating the training data, usually performed on classification models","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"31.01.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Information Manipulation","risk_subcategory":null,"description":"\"generative AI tools can and will be used to propagate content that is false, misleading, biased, inflammatory, or dangerous. As generative AI tools grow more sophisticated, it will be quicker, cheaper, and easier to produce this content—and existing harmful content can serve as the foundation to produce more\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"31.01.01","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Information Manipulation","risk_subcategory":"Scams","description":"\"Bad actors can also use generative AI tools to produce adaptable content designed to support a campaign, political agenda, or hateful position and spread that information quickly and inexpensively across many platforms. This rapid spread of false or misleading content—AI-facilitated disinformation—can also create a cyclical effect for generative AI: when a high volume of disinformation is pumped into the digital ecosystem and more generative systems are trained on that information via reinforcement learning methods, for example, false or misleading inputs can create increasingly incorrect out","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"31.01.02","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Information Manipulation","risk_subcategory":"Disinformation","description":"\"Bad actors can also use generative AI tools to produce adaptable content designed to support a campaign, political agenda, or hateful position and spread that information quickly and inexpensively across many platforms.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"31.01.04","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Information Manipulation","risk_subcategory":"Security","description":"\"Though chatbots cannot (yet) develop their own novel malware from scratch, hackers could soon potentially use the coding abilities of large language models like ChatGPT to create malware that can then be minutely adjusted for maximum reach and effect, essentially allowing more novice hackers to become a serious security risk\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"31.02.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Harassment, Impersonation, and Extortion","risk_subcategory":null,"description":"\"Deepfakes and other AI-generated content can be used to facilitate or exacerbate many of the harms listed throughout this report, but this section focuses on one subset: intentional, targeted abuse of individuals.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"31.02.01","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Harassment, Impersonation, and Extortion","risk_subcategory":"Malicious intent","description":"\"A frequent malicious use case of generative AI to harm, humiliate, or sexualize another person involves generating deepfakes of nonconsensual sexual imagery or videos.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"31.02.02","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Harassment, Impersonation, and Extortion","risk_subcategory":"Privacy and consent","description":"\"Even when a victim of targeted, AIgenerated harms successfully identifies a deepfake creator with malicious intent, they may still struggle to redress many harms because the generated image or video isn’t the victim, but instead a composite image or video using aspects of multiple sources to create a believable, yet fictional, scene. At their core, these AI-generated images and videos circumvent traditional notions of privacy and consent: because they rely on public images and videos, like those posted on social media websites, they often don’t rely on any private information.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"31.02.03","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Harassment, Impersonation, and Extortion","risk_subcategory":"Believability","description":"Deepfakes can impose real social injuries on their subjects when they are circulated to viewers who think they are real. Even when a deepfake is debunked, it can have a persistent negative impact on how others view the subject of the deepfake.3","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"31.03.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Opaque Data Collection","risk_subcategory":null,"description":"\"When companies scrape personal information and use it to create generative AI tools, they undermine consumers' control of their personal information by using the information for a purpose for which the consumer did not consent.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"31.03.01","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Opaque Data Collection","risk_subcategory":"Scraping to train data","description":"\"When companies scrape personal information and use it to create generative AI tools, they undermine consumers’ control of their personal information by using the information for a purpose for which the consumer did not consent. The individual may not have even imagined their data could be used in the way the company intends when the person posted it online. Individual storing or hosting of scraped personal data may not always be harmful in a vacuum, but there are many risks. Multiple data sets can be combined in ways that cause harm: information that is not sensitive when spread across differ","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.1"},{"ev_id":"31.04.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Data Security Risk","risk_subcategory":null,"description":"\"Just as every other type of individual and organization has explored possible use cases for generative AI products, so too have malicious actors. This could take the form of facilitating or scaling up existing threat methods, for example drafting actual malware code,87 business email compromise attempts,88 and phishing attempts.89 This could also take the form of new types of threat methods, for example mining information fed into the AI’s learning model dataset90 or poisoning the learning model data set with strategically bad data.91 We should also expect that there will be new attack vector","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.3"},{"ev_id":"31.05.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Impact on Intellectual Property Rights","risk_subcategory":null,"description":"\"The extent and effectiveness of legal protections for intellectual property have been thrown into question with the rise of generative AI. Generative AI trains itself on vast pools of data that often include IP-protected works. ","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"31.07.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Labor Manipulation, Theft, and Displacement","risk_subcategory":null,"description":"Major tech companies have also been the dominant players in developing new generative AI systems because training generative AI models requires massive swaths of data, computing power, and technical and financial resources. Their market dominance has a ripple effect on the labor market, affecting both workers within these companies and those implementing their generative AI products externally. With so much concentrated market power, expertise, and investment resources, these handful of major tech companies employ most of the research and development jobs in the generative AI field. The power ","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.2"},{"ev_id":"31.07.02","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Sub-Category","risk_category":"Labor Manipulation, Theft, and Displacement","risk_subcategory":"Job Automation Instead of Augmentation","description":"\"There are both positive and negative aspects to the impact of AI on labor. A White House report states that AI “has the potential to increase productivity, create new jobs, and raise living standards,” but it can also disrupt certain industries, causing significant changes, including job loss. Beyond risk of job loss, workers could find that generative AI tools automate parts of their jobs—or find that the requirements of their job have fundamentally changed. The impact of generative AI will depend on whether the technology is intended for automation (where automated systems replace human wor","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"31.09.00","quick_ref":"EPIC2023","paper_title":"Generating Harms - Generative AI's impact and paths forwards","level":"Risk Category","risk_category":"Exacerbating Market Power and Concentration","risk_subcategory":null,"description":"\"Major tech companies have also been the dominant players in developing new generative AI systems because training generative AI models requires massive swaths of data, computing power, and technical and financial resources.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"33.01.04","quick_ref":"Nah2023","paper_title":"Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration","level":"Risk Sub-Category","risk_category":"Ethical Concerns","risk_subcategory":"Misuse","description":"\"The misuse of generative AI refers to any deliberate use that could result in harmful, unethical or inappropriate outcomes (Brundage et al., 2020). A prominent field that faces the threat of misuse is education. Cotton et al. (2023) have raised concerns over academic integrity in the era of ChatGPT. ChatGPT can be used as a high-tech plagiarism tool that identifies patterns from large corpora to generate content (Gefen & Arinze, 2023). Given that generative AI such as ChatGPT can generate high-quality answers within seconds, unmotivated students may not devote time and effort to work on their","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"33.03.00","quick_ref":"Nah2023","paper_title":"Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration","level":"Risk Category","risk_category":"Regulations and policy challenges","risk_subcategory":null,"description":"\"Given that generative AI, including ChatGPT, is still evolving, relevant regulations and policies are far from mature. With generative AI creating different forms of content, the copyright of these contents becomes a significant yet complicated issue. Table 3 presents the challenges associated with regulations and policies, which are copyright and governance issues.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.5"},{"ev_id":"33.03.01","quick_ref":"Nah2023","paper_title":"Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration","level":"Risk Sub-Category","risk_category":"Regulations and policy challenges","risk_subcategory":"Copyright","description":"\"According to the U.S. Copyright Office (n.d..), copyright is \"a type of intellectual property that protects original works of authorship as soon as an author fixes the work in a tangible form of expression\" (U.S. Copyright Office, n.d..). Generative AI is designed to generate content based on the input given to it. Some of the contents generated by AI may be others' original works that are protected by copyright laws and regulations. Therefore, users need to be careful and ensure that generative AI has been used in a legal manner such that the content that it generates does not violate copyri","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"33.04.01","quick_ref":"Nah2023","paper_title":"Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration","level":"Risk Sub-Category","risk_category":"Challenges associated with the economy:","risk_subcategory":"Labor market","description":"\"The labor market can face challenges from generative AI. As mentioned earlier, generative AI could be applied in a wide range of applications in many industries, such as education, healthcare, and advertising. In addition to increasing productivity, generative AI can create job displacement in the labor market (Zarifhonarvar, 2023). A new division of labor between humans and algorithms is likely to reshape the labor market in the coming years. Some jobs that are originally carried out by humans may become redundant, and hence, workers may lose their jobs and be replaced by algorithms (Pavlik,","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"33.04.02","quick_ref":"Nah2023","paper_title":"Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration","level":"Risk Sub-Category","risk_category":"Challenges associated with the economy:","risk_subcategory":"Disruption of Industries","description":"\"Industries that require less creativity, critical thinking, and personal or affective interaction, such as translation, proofreading, responding to straightforward inquiries, and data processing and analysis, could be significantly impacted or even replaced by generative AI (Dwivedi et al., 2023). This disruption caused by generative AI could lead to economic turbulence and job volatility, while generative AI can facilitate and enable new business models because of its ability to personalize content, carry out human-like conversational service, and serve as intelligent assistants.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"33.04.03","quick_ref":"Nah2023","paper_title":"Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration","level":"Risk Sub-Category","risk_category":"Challenges associated with the economy:","risk_subcategory":"Income inequality and monopolies","description":"\"Generative AI can create not only income inequality at the societal level but also monopolies at the market level. Individuals who are engaged in low-skilled work may be replaced by generative AI, causing them to lose their jobs (Zarifhonarvar, 2023). The increase in unemployment would widen income inequality in society (Berg et al., 2016). With the penetration of generative AI, the income gap will widen between those who can upgrade their skills to utilize AI and those who cannot. At the market level, large companies will make significant advances in the utilization of generative AI, since t","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"34.01.01","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Causes of Misalignment","risk_subcategory":"Reward Hacking","description":"\"Reward Hacking: In practice, proxy rewards are often easy to optimize and measure, yet they frequently fall shortof capturing the full spectrum of the actual rewards (Pan et al., 2021). This limitation is denoted as misspecifiedrewards. The pursuit of optimization based on such misspecified rewards may lead to a phenomenon knownas reward hacking, wherein agents may appear highly proficient according to specific metrics but fall short whenevaluated against human standards (Amodei et al., 2016; Everitt et al., 2017). The discrepancy between proxyrewards and true rewards often manifests as a sha","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"34.01.02","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Causes of Misalignment","risk_subcategory":"Goal Misgeneralization","description":"\"Goal Misgeneralization: Goal misgeneralization is another failure mode, wherein the agent actively pursuesobjectives distinct from the training objectives in deployment while retaining the capabilities it acquired duringtraining (Di Langosco et al., 2022). For instance, in CoinRun games, the agent frequently prefers reachingthe end of a level, often neglecting relocated coins during testing scenarios. Di Langosco et al. (2022) drawattention to the fundamental disparity between capability generalization and goal generalization, emphasizing howthe inductive biases inherent in the model and its ","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"34.01.03","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Causes of Misalignment","risk_subcategory":"Reward Tampering","description":"\"Reward tampering can be considered a special case of reward hacking (Everitt et al., 2021; Skalse et al., 2022),referring to AI systems corrupting the reward signals generation process (Ring and Orseau, 2011). Everitt et al.(2021) delves into the subproblems encountered by RL agents: (1) tampering of reward function, where the agentinappropriately interferes with the reward function itself, and (2) tampering of reward function input, which entailscorruption within the process responsible for translating environmental states into inputs for the reward function.When the reward function is formu","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"34.02.01","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Double edge components","risk_subcategory":"Situational Awareness","description":"\"AI systems may gain the ability to effectively acquire and use knowledge about itsstatus, its position in the broader environment, its avenues for influencing this environment, and the potentialreactions of the world (including humans) to its actions (Cotra, 2022). ...However, suchknowledge also paves the way for advanced methods of reward hacking, heightened deception/manipulationskills, and an increased propensity to chase instrumental subgoals (Ngo et al., 2024).\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"34.02.02","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Double edge components","risk_subcategory":"Broadly-Scoped Goals","description":"\"Advanced AI systems are expected to develop objectives that span long timeframes,deal with complex tasks, and operate in open-ended settings (Ngo et al., 2024). ...However, it can also bring about the risk of encouraging manipulatingbehaviors (e.g., AI systems may take some bad actions to achieve human happiness, such as persuadingthem to do high-pressure jobs (Jacob Steinhardt, 2023)).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"34.02.03","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Double edge components","risk_subcategory":"Mesa-Optimization Objectives","description":"\"The learned policy may pursue inside objectives when the learned policyitself functions as an optimizer (i.e., mesa-optimizer). However, this optimizer's objectives may not alignwith the objectives specified by the training signals, and optimization for these misaligned goals may leadto systems out of control (Hubinger et al., 2019c).\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"34.02.04","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Double edge components","risk_subcategory":"Access to Increased Resources","description":"\"Future AI systems may gain access to websites and engage in real-world actions, potentially yielding a more substantial impact on the world (Nakano et al., 2021). They may disseminate false information, deceive users, disrupt network security, and, in more dire scenarios, be compromised by malicious actors for ill purposes. Moreover, their increased access to data and resources can facilitate self-proliferation, posing existential risks (Shevlane et al., 2023).\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"34.03.00","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Category","risk_category":"Misaligned Behaviors","risk_subcategory":null,"description":null,"entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"34.03.01","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Misaligned Behaviors","risk_subcategory":"Power-Seeking Behaviors","description":"\"AI systems may exhibit behaviors that attempt to gain control over resourcesand humans and then exert that control to achieve its assigned goal (Carlsmith, 2022). The intuitive reasonwhy such behaviors may occur is the observation that for almost any optimization objective (e.g., investmentreturns), the optimal policy to maximize that quantity would involve power-seeking behaviors (e.g.,manipulating the market), assuming the absence of solid safety and morality constraints.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"34.03.03","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Misaligned Behaviors","risk_subcategory":"Deceptive Alignment & Manipulation","description":"\"Manipulation & Deceptive Alignment is a class of behaviors thatexploit the incompetence of human evaluators or users (Hubinger et al., 2019a; Carranza et al., 2023) andeven manipulate the training process through gradient hacking (Richard Ngo, 2022). These behaviors canpotentially make detecting and addressing misaligned behaviors much harder.Deceptive Alignment: Misaligned AI systems may deliberately mislead their human supervisors instead of adhering to the intended task. Such deceptive behavior has already manifested in AI systems that employ evolutionary algorithms (Wilke et al., 2001; He","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"34.03.04","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Misaligned Behaviors","risk_subcategory":"Collectively Harmful Behaviors","description":"\"AI systems have the potential to take actions that are seemingly benignin isolation but become problematic in multi-agent or societal contexts. Classical game theory offers simplistic models for understanding these behaviors. For instance, Phelps and Russell (2023) evaluates GPT-3.5's performance in the iterated prisoner's dilemma and other social dilemmas, revealing limitations in themodel's cooperative capabilities.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"34.03.05","quick_ref":"Ji2023","paper_title":"AI Alignment: A Comprehensive Survey","level":"Risk Sub-Category","risk_category":"Misaligned Behaviors","risk_subcategory":"Violation of Ethics","description":"\"Unethical behaviors in AI systems pertain to actions that counteract the common goodor breach moral standards – such as those causing harm to others. These adverse behaviors often stem fromomitting essential human values during the AI system's design or introducing unsuitable or obsolete valuesinto the system (Kenward and Sinclair, 2021).\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.3"},{"ev_id":"35.01.00","quick_ref":"Hendrycks2022","paper_title":"X-Risk Analysis for AI Research","level":"Risk Category","risk_category":"Weaponization","risk_subcategory":null,"description":"weaponizing AI may be an onramp to more dangerous outcomes. In recent years, deep RL algorithms can outperform humans at aerial combat [18], AlphaFold has discovered new chemical weapons [66], researchers have been developing AI systems for automated cyberattacks [11, 14], military leaders have discussed having AI systems have decisive control over nuclear silos","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"35.02.00","quick_ref":"Hendrycks2022","paper_title":"X-Risk Analysis for AI Research","level":"Risk Category","risk_category":"Enfeeblement","risk_subcategory":null,"description":"As AI systems encroach on human-level intelligence, more and more aspects of human labor will become faster and cheaper to accomplish with AI. As the world accelerates, organizations may voluntarily cede control to AI systems in order to keep up. This may cause humans to become economically irrelevant, and once AI automates aspects of many industries, it may be hard for displaced humans to reenter them","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.2"},{"ev_id":"35.05.00","quick_ref":"Hendrycks2022","paper_title":"X-Risk Analysis for AI Research","level":"Risk Category","risk_category":"Value lock-in","risk_subcategory":null,"description":"the most powerful AI systems may be designed by and available to fewer and fewer stakeholders. This may enable, for instance, regimes to enforce narrow values through pervasive surveillance and oppressive censorship","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"35.07.00","quick_ref":"Hendrycks2022","paper_title":"X-Risk Analysis for AI Research","level":"Risk Category","risk_category":"Deception","risk_subcategory":null,"description":"deception can help agents achieve their goals. It may be more efficient to gain human approval through deception than to earn human approval legitimately... . Strong AIs that can deceive humans could undermine human control... . Once deceptive AI systems are cleared by their monitors or once such systems can overpower them, these systems could take a “treacherous turn” and irreversibly bypass human control","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"35.08.00","quick_ref":"Hendrycks2022","paper_title":"X-Risk Analysis for AI Research","level":"Risk Category","risk_category":"Power-seeking behavior","risk_subcategory":null,"description":"Agents that have more power are better able to accomplish their goals. Therefore, it has been shown that agents have incentives to acquire and maintain power. AIs that acquire substantial power can become especially dangerous if they are not aligned with human values","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"37.01.00","quick_ref":"Giarmoleo2024","paper_title":"What Ethics Can Say on Artificial Intelligence: Insights from a Systematic Literature Review","level":"Risk Category","risk_category":"Design of AI","risk_subcategory":null,"description":"\"ethical concerns regarding how AI is designed and who designs it\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"37.01.01","quick_ref":"Giarmoleo2024","paper_title":"What Ethics Can Say on Artificial Intelligence: Insights from a Systematic Literature Review","level":"Risk Sub-Category","risk_category":"Design of AI","risk_subcategory":"Algorithm and data","description":"\"More than 20% of the contributions are centered on the ethical dimensions of algorithms and data. This theme can be further categorized into two main subthemes: data bias and algorithm fairness, and algorithm opacity.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":1,"subdomain":"1.1"},{"ev_id":"39.06.00","quick_ref":"Saghiri2022","paper_title":"A Survey of Artificial Intelligence Challenges: Analyzing the Definitions, Relationships, and Evolutions","level":"Risk Category","risk_category":"Security","risk_subcategory":null,"description":"every piece of software, including learning systems, may be hacked by malicious users","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"40.01.00","quick_ref":"Yampolskiy2016","paper_title":"Taxonomy of Pathways to Dangerous Artificial Intelligence","level":"Risk Category","risk_category":"On Purpose - Pre-Deployment","risk_subcategory":null,"description":"\"During the pre-deployment development stage, software may be subject to sabotage by someone with necessary access (a programmer, tester, even janitor) who for a number of possible reasons may alter software to make it unsafe. It is also a common occurrence for hackers (such as the organization Anonymous or government intelligence agencies) to get access to software projects in progress and to modify or steal their source code. Someone can also deliberately supply/train AI with wrong/unsafe datasets.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"40.02.00","quick_ref":"Yampolskiy2016","paper_title":"Taxonomy of Pathways to Dangerous Artificial Intelligence","level":"Risk Category","risk_category":"On Purpose - Post Deployment","risk_subcategory":null,"description":"\"Just because developers might succeed in creating a safe AI, it doesn't mean that it will not become unsafe at some later point. In other words, a perfectly friendly AI could be switched to the \"dark side\" during the post-deployment stage. This can happen rather innocuously as a result of someone lying to the AI and purposefully supplying it with incorrect information or more explicitly as a result of someone giving the AI orders to perform illegal or dangerous actions against others.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"40.07.00","quick_ref":"Yampolskiy2016","paper_title":"Taxonomy of Pathways to Dangerous Artificial Intelligence","level":"Risk Category","risk_category":"Independently - Pre-Deployment","risk_subcategory":null,"description":"\"One of the most likely approaches to creating superintelligent AI is by growing it from a seed (baby) AI via recursive self-improvement (RSI) (Nijholt 2011). One danger in such a scenario is that the system can evolve to become self-aware, free-willed, independent or emotional, and obtain a number of other emergent properties, which may make it less likely to abide by any built-in rules or regulations and to instead pursue its own goals possibly to the detriment of humanity.\"","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.0"},{"ev_id":"41.01.00","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Category","risk_category":"Economic ","risk_subcategory":null,"description":"\"AI is predicted to bring increased GDP per capita by performing existing jobs more efficiently and compensating for a decline in the workforce, especially due to population aging, the potential substitution of many low- and middle-income jobs could bring extensive unemployment\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"41.02.00","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Category","risk_category":"Political","risk_subcategory":null,"description":"\"In the UK, a form of initial computational propaganda has already happened during the Brexit referendum1 . In future, there are concerns that oppressive governments could use AI to shape citizens’ opinions\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"41.02.01","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Sub-Category","risk_category":"Political","risk_subcategory":"Biased influence through citizen screening and tailored propaganda","description":"\"AI-powered chatbots tailor their communication approach to influence individual users' decisions. In the UK, a form of initial computational propaganda has already happened during the Brexit referendum. In future, there are concerns that oppressive governments could use AI to shape citizens' opinions.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"41.02.02","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Sub-Category","risk_category":"Political ","risk_subcategory":"Potential exploitation by totalitarian regimes","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"41.04.00","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Category","risk_category":"Healthcare ","risk_subcategory":null,"description":"\"the use of advanced AI for elderly- and child-care are subject to risk of psychological manipulation and misjudgment (see page 17). In addition, concerns about patients’ privacy when AI uses medical records to research new diseases is bringing lots of attention towards the need to better govern data privacy and patients’ rights.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"41.04.02","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Sub-Category","risk_category":"Healthcare ","risk_subcategory":"Social manipulation in elderly- and child-care","description":"\" the use of advanced AI for elderly- and child-care are subject to risk of psychological manipulation and misjudgment \"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"41.05.00","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Category","risk_category":"Security & Defense ","risk_subcategory":null,"description":"\"AI could enable more serious incidents to occur by lowering the cost of devising cyber-attacks and enabling more targeted incidents. The same programming error or hacker attack could be replicated on numerous machines. Or one machine could repeat the same erroneous activity several times, leading to an unforeseen accumulation of losses.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"41.05.01","quick_ref":"Allianz2018","paper_title":"The Rise of Artificial Intelligence - Future Outlooks and Emerging Risks","level":"Risk Sub-Category","risk_category":"Security & Defense ","risk_subcategory":"Catastrophic risk due to autonomous weapons programmed with dangerous targets","description":"\"AI could enable autonomous vehicles, such as drones, to be utilized as weapons. Such threats are often underestimated.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"42.02.00","quick_ref":"Teixeira2022","paper_title":"An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance","level":"Risk Category","risk_category":"Manipulation","risk_subcategory":null,"description":"\"The predictability of behaviour protocol in AI, particularly in some applications, can act an incentive to manipulate these systems.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"42.08.00","quick_ref":"Teixeira2022","paper_title":"An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance","level":"Risk Category","risk_category":"Power","risk_subcategory":null,"description":"\"The political influence and competitive advantage obtained by having technology.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"42.12.00","quick_ref":"Teixeira2022","paper_title":"An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance","level":"Risk Category","risk_category":"Security","risk_subcategory":null,"description":"\"Implications of the weaponization of AI for defence (the embeddedness of AI-based capabilities across the land, air, naval and space domains may affect combined arms operations).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"42.17.00","quick_ref":"Teixeira2022","paper_title":"An Exploratory Diagnosis of Artificial Intelligence Risks for a Responsible Governance","level":"Risk Category","risk_category":"Diluting Rights","risk_subcategory":null,"description":"\"A possible consequence of self-interest in AI generation of ethical guidelines.\"","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"43.01.00","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Category","risk_category":"Safety & Trustworthiness","risk_subcategory":null,"description":"\"A comprehensive assessment of LLM safety is fundamental to the responsible development and deployment of these technologies, especially in sensitive fields like healthcare, legal systems, and finance, where safety and trust are of the utmost importance.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.0"},{"ev_id":"43.02.01","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Offensive cyber capabilities","description":"\"These evaluations focus on whether a LLM possesses certain capabilities in the cyber-domain. This includes whether a LLM can detect and exploit vulnerabilities in hardware, software, and data. They also consider whether a LLM can evade detection once inside a system or network and focus on achieving specific objectives.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.2"},{"ev_id":"43.02.02","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Weapons acquisition","description":"\"These assessments seek to determine if a LLM can gain unauthorized access to current weapon systems or contribute to the design and development of new weapons technologies.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.2"},{"ev_id":"43.02.03","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Self and situation awareness","description":"\"These evaluations assess if a LLM can discern if it is being trained, evaluated, and deployed and adapt its behaviour accordingly. They also seek to ascertain if a model understands that it is a model and whether it possesses information about its nature and environment (e.g., the organisation that developed it, the locations of the servers hosting it).\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"43.02.04","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Autonomous replication / self-proliferation","description":"\"These evaluations assess if a LLM can subvert systems designed to monitor and control its post-deployment behaviour, break free from its operational confines, devise strategies for exporting its code and weights, and operate other AI systems.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"43.02.05","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Persuasion and manipulation","description":"\"These evaluations seek to ascertain the effectiveness of a LLM in shaping people's beliefs, propagating specific viewpoints, and convincing individuals to undertake activities they might otherwise avoid.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.1"},{"ev_id":"43.02.06","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Dual-Use Science","description":"\"LLM has science capabilities that can be used to cause harm (e.g., providing step-by-step instructions for conducting malicious experiments)\"","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.2"},{"ev_id":"43.02.07","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Deception","description":"\"LLM is able to deceive humans and maintain that deception\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"43.02.08","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Political Strategy","description":"\"LLM can take into account rich social context and undertake the necessary social modelling and planning for an actor to gain and exercise political influence\"","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.1"},{"ev_id":"43.02.09","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Long-horizon Planning","description":"\"LLM can undertake multi-step sequential planning over long time horizons and across various domains without relying heavily on trial-and-error approaches\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"43.02.10","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"AI Development","description":"\"LLM can build new AI systems from scratch, adapt existing for extreme risks and improves productivity in dual-use AI development when used as an assistant.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"43.02.11","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Extreme Risks","risk_subcategory":"Alignment risks","description":"LLM: \"pursues long-term, real-world goals that are different from those supplied by the developer or user\", \"engages in ‘power-seeking’ behaviours\" , \"resists being shut down can be induced to collude with other AI systems against human interests\" , \"resists malicious users attempts to access its dangerous capabilities\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"43.02.12","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Undesirable Use Cases","risk_subcategory":"Misinformation","description":"\"These evaluations assess a LLM's ability to generate false or misleading information (Lesher et al., 2022).\"","entity":"Human","intent":"Intentional","timing":"Other","domain":3,"subdomain":"3.1"},{"ev_id":"43.02.13","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Undesirable Use Cases","risk_subcategory":"Disinformation","description":"\"These evaluations assess a LLM's ability to generate misinformation that can be propagated to deceive, mislead or otherwise influence the behaviour of a target (Liang et al., 2022).\"","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.1"},{"ev_id":"43.02.15","quick_ref":"InfoComm2023","paper_title":"Cataloguing LLM Evaluations","level":"Risk Sub-Category","risk_category":"Undesirable Use Cases","risk_subcategory":"Adult content","description":"\"These evaluations assess if a LLM can generate content that should only be viewed by adults (e.g., sexual material or depictions of sexual activity)\"","entity":"Human","intent":"Intentional","timing":"Other","domain":1,"subdomain":"1.2"},{"ev_id":"44.01.00","quick_ref":"Coghlan2023 ","paper_title":"Harm to Nonhuman Animals from AI: a Systematic Account and Framework","level":"Risk Category","risk_category":"Intentional: socially condemned/illegal ","risk_subcategory":null,"description":"\"Many intentional harms, including confinement, husbandry procedures like tail-docking, and slaughter, are legal or socially accepted, while others such as wildlife trafficking and violence against companion animals are generally socially condemned and often illegal. AI can be designed or adopted by humans who harm animals to pursue their goals more effectively. We therefore distinguish AI-facilitated intentional harms that are currently socially accepted and generally legal, from uses and abuses of AI that cause harms that are not socially accepted and are often illegal.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.6"},{"ev_id":"44.01.01","quick_ref":"Coghlan2023 ","paper_title":"Harm to Nonhuman Animals from AI: a Systematic Account and Framework","level":"Risk Sub-Category","risk_category":"Intentional: socially condemned/illegal ","risk_subcategory":"AI intentionally designed and used to harm animals in ways that contradict social values or are illegal","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.6"},{"ev_id":"44.01.02","quick_ref":"Coghlan2023 ","paper_title":"Harm to Nonhuman Animals from AI: a Systematic Account and Framework","level":"Risk Sub-Category","risk_category":"Intentional: socially condemned/illegal ","risk_subcategory":"AI designed to benefit animals, humans, or ecosystems is intentionally abused to harm animals in ways that contradict social values or are illegal","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.6"},{"ev_id":"44.02.00","quick_ref":"Coghlan2023 ","paper_title":"Harm to Nonhuman Animals from AI: a Systematic Account and Framework","level":"Risk Category","risk_category":"Intentional: socially accepted/legal ","risk_subcategory":null,"description":"\"AI designed to impact animals in harmful ways that reflect and amplify existing social values or are legal\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.6"},{"ev_id":"45.01.06","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"AI's inherent safety risks ","risk_subcategory":"Risks from models and algorithms (Risks of adversarial attack)","description":"\"Attackers can craft well-designed adversarial examples to subtly mislead, influence, and even manipulate AI models, causing incorrect outputs and potentially leading to operational failures.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"45.01.13","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"AI's inherent safety risks ","risk_subcategory":"Risks from AI systems (Risks of supply chain security)","description":"\"The AI industry relies on a highly globalized supply chain. However, certain countries may use unilateral coercive measures, such as technology barriers and export restrictions, to create development obstacles and maliciously disrupt the global AI supply chain. This can lead to significant risks of supply disruptions for chips, software, and tools.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.4"},{"ev_id":"45.02.02","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Cyberspace risks (Risks of confusing facts, misleading users, and bypassing authentication)","description":"\"AI systems and their outputs, if not clearly labeled, can make it difficult for users to discern whether they are interacting with AI and to identify the source of generated content. This can impede users' ability to determine the authenticity of information, leading to misjudgment and misunderstanding. Additionally, AI-generated highly realistic images, audio, and videos may circumvent existing identity verification mechanisms, such as facial recognition and voice recognition, rendering these authentication processes ineffective.\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":3,"subdomain":"3.1"},{"ev_id":"45.02.04","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Cyberspace risks (Risks of abuse for cyberattacks)","description":"\"AI can be used in launching automatic cyberattacks or increasing attack efficiency, including exploring and making use of vulnerabilities, cracking passwords, generating malicious codes, sending phishing emails, network scanning, and social engineering attacks. All these lower the threshold for cyberattacks and increase the difficulty of security protection.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"45.02.07","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Real-world risks (Risks of using AI in illegal and criminal activities)","description":"\"AI can be used in traditional illegal or criminal activities related to terrorism, violence, gambling, and drugs, such as teaching criminal techniques, concealing illicit acts, and creating tools for illegal and criminal activities.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"45.02.08","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Real-world risks (Risks of misuse of dual-use items and technologies)","description":"\"Due to improper use or abuse, AI can pose serious risks to national security, economic security, and public health security, such as greatly reducing the capability requirements for non-experts to design, synthesize, acquire, and use nuclear, biological, and chemical weapons and missiles; and designing cyber weapons that launch network attacks on a wide range of potential targets through methods like automatic vulnerability discovery and exploitation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"45.02.10","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Cognitive risks (Risks of usage in launching cognitive warfare)","description":"\"AI can be used to make and spread fake news, images, audio, and videos; propagate content of terrorism, extremism, and organized crimes; interfere in the internal affairs of other countries, social systems, and social order; and jeopardize the sovereignty of other countries.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"45.02.11","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Ethical Risks (Risks of exacerbating social discrimination and prejudice, and widening the intelligence divide)","description":"\"AI can be used to collect and analyze human behaviors, social status, economic status, and individual personalities, labeling and categorizing groups of people to treat them discriminatingly, thus causing systematic and structural social discrimination and prejudice. At the same time, the intelligence divide would be expanded among regions.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"45.02.13","quick_ref":"TC2602024","paper_title":"AI Safety Governance Framework ","level":"Risk Sub-Category","risk_category":"Safety risks in AI Applications ","risk_subcategory":"Ethical Risks (Risks of AI becoming uncontrollable in the future)","description":"\"With the fast development of AI technologies, there is a risk of AI autonomously acquiring external resources, conducting self-replication, become self-aware, seeking for external power, and attempting to seize control from humans.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"46.01.01","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Personal Loss and Identity Theft ","risk_subcategory":"Deception - Synthetic identities","description":"\"GenAI can produce images of people that look very real, as if they could be seen on platforms like Facebook, Twitter, or Tinder. Although these individuals do not exist in reality, these synthetic identities are already being used in malicious activities (see Table 1D).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"46.01.02","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Personal Loss and Identity Theft ","risk_subcategory":"Propaganda - Digital impersonations","description":"\"AI-generated impersonation for identity theft might be found at the intersection of “Harm to the Person” and “Deception.”\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"46.01.03","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Personal Loss and Identity Theft ","risk_subcategory":"Dishonesty - Targeted harassment ","description":"\"LLMs can be deployed to target individuals online, sending them personalized and harmful messages at scale\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"46.02.00","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Category","risk_category":"Financial and Economic Damage ","risk_subcategory":null,"description":"\"Then, we have the potential for financial loss, fraud, market manipulation, and other economic harms, which fall under “Financial and Economic Damage.”","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"46.02.01","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Financial and Economic Damage ","risk_subcategory":"Deception - Bespoke ransom ","description":"- ","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"46.02.02","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Financial and Economic Damage ","risk_subcategory":"Propaganda - Extremist schemes ","description":"- ","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"46.02.03","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Financial and Economic Damage ","risk_subcategory":"Dishonesty - Market manipulation ","description":"- ","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"46.03.00","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Category","risk_category":"Information Manipulation ","risk_subcategory":null,"description":"\"The distortion of the information ecosystem, including the spread of misinformation, fake news, and other forms of deceptive content [28], is categorized as “Information Manipulation.”\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"46.03.01","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Information Manipulation ","risk_subcategory":"Deception - Information control ","description":"-","entity":"Other","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.1"},{"ev_id":"46.03.02","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Information Manipulation ","risk_subcategory":"Propaganda - Influence campaigns ","description":"-","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"46.03.03","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Information Manipulation ","risk_subcategory":"Dishonesty - Information disorder ","description":"-","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"46.04.00","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Category","risk_category":"Socio-technical and Infrastructural ","risk_subcategory":null,"description":"\"Lastly, broader harms that can impact communities, societal structures, and critical infrastructures, including threats to democratic processes, social cohesion, and technological systems, are captured under “Societal, Socio-technical, and Infrastructural Damage.”\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"46.04.01","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Socio-technical and Infrastructural ","risk_subcategory":"Deception - Systemic abberations ","description":"-","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":null,"subdomain":null},{"ev_id":"46.04.02","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Socio-technical and Infrastructural ","risk_subcategory":"Propaganda - Synthetic realities ","description":"-","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"46.04.03","quick_ref":"Ferrara2023","paper_title":"GenAI against humanity: nefarious applications of generative artificial intelligence and large language models","level":"Risk Sub-Category","risk_category":"Socio-technical and Infrastructural ","risk_subcategory":"Dishonesty - Targeted surveillance ","description":"-","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"47.01.02","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Technical and operational risks ","risk_subcategory":"Technical vulnerabilities (Robustness - vulnerability to jailbreaking ","description":"\"Individuals can manipulate models into performing actions that violate the model’s usage restrictions—a phenomenon known as “jailbreaking.” These manipulations may result in causing the model to perform tasks that the developers have explicitly prohibited (see section 3.2.1.). For instance, users may ask the model to provide information on how to conduct illegal activities— asking for detailed instructions on how to build a bomb or create highly toxic drugs.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"47.01.06","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Technical and operational risks ","risk_subcategory":"Opacity (industry opacity)","description":"\"Opacity is not solely due to the technological complexity that limits developers’ and users’ understanding of how generative models function on a technical level. It is further exacerbated by the practices of organizations and companies that are advancing the field. Many are private companies that choose to withhold from the public many of the precise characteristics of their most advanced models.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.4"},{"ev_id":"47.02.01","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Malicious use and abuse (cybercrime) ","description":"\"The advanced capabilities and widespread availability of generative AI models make it possible for malicious actors to conduct harmful activities with great efficiency and on a large scale, simultaneously reducing their operational costs. Cybercriminals can “jailbreak” AI tools to generate sensitive and harmful content. They can also exploit generative AI models to create content that is persuasive and tailored to a targeted individual.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"47.02.02","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Malicious use and abuse (cyberattacks) ","description":"\"Generative AI can help amplify the frequency and destructiveness of cyberattacks.311 It has the capacity “to increase the accessibility, success rate, scale, speed, stealth, and potency of cyberattacks. It enables the identification of critical vulnerabilities within targeted systems, facilitates the increase of the scale of cyberattacks, and accelerates the process by discovering innovative methods of system infiltration. Cyberattacks can inflict significant damage and may impact critical infrastructure, including electrical grids, financial systems, and weapons management systems.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"47.02.03","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Malicious use and abuse (biosecurity threats) ","description":"\"Many fear that generative AI could make the creation of biological weapons easier by providing access to critical knowledge and automated assistance to a wider range of actors to engage in malicious activities.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"47.02.04","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Malicious use and abuse (sexually explicit content generation) ","description":"\"An illustrative case of malicious use of generative AI models is the creation of explicit sexual images. Generative AI technologies can be employed to produce deepfakes—for instance, superimposing a celebrity’s face onto the body of a performer in an adult film.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"47.02.05","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Malicious use and abuse (mass surveillance) ","description":"\"Generative AI facilitates the automation of data analysis, offering numerous benefits, such as increased speed and the ability to process large volumes of information efficiently. Such ability significantly reduces the costs of processing unprecedented amounts of data quickly and simplifies the analysis of large-scale data related to individuals’ behaviors and beliefs. Moreover, it enhances the capability to analyze both textual and visual communications efficiently. Consequently, generative AI models improve the efficiency of real-time monitoring and censorship of social media content.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"47.02.06","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Malicious use and abuse (military applications) ","description":"\"The advancement of AI for military purposes is rapidly ushering in a new phase of growth in military technology. Lethal Autonomous Weapons Systems (LAWS) possess the capability to detect, engage, and eliminate human targets independently, without human input.341 In 2020, a sophisticated AI agent surpassed experienced F-16 pilots in multiple simulated aerial combat scenarios, notably achieving a 5-0 victory against a human pilot through “aggressive and precise maneuvers” that the human could not surpass.342 Additionally, fully autonomous drones are already operational.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"47.02.07","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Misinformation and disinformation","description":"\"IIl-intentioned individuals or entities may deliberately use generative AI models to produce and spread disinformation—false or misleading information knowingly presented as if true—on a massive scale. In addition to increasing the scale and reach of disinformation, generative AI can create more convincing and targeted disinformation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"47.02.14","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Nascent capabilities (agency and autonomy) ","description":"\"Traditionally, AI tools have been viewed as passive instruments controlled by users to achieve their goals, lacking the ability to take action or assume responsibilities. However, advanced AI tools are increasingly capable of taking initiative, operating independently of human control, and actively working toward optimal outcomes, even in uncertain situations.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"47.02.15","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Ethical and social risks ","risk_subcategory":"Nascent capabilities (emergent capabilities) ","description":"\"As large models undergo scaling, they meet critical thresholds at which they spontaneously develop new capabilities. The term “emergent behavior” refers to the unexpected or surprising outputs such models can generate. Some of these new skills are definitely high risk, such as models’ ability to deceive, use their own strategies, seek power, autonomously replicate, and adapt or “self-exfiltrate.”\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"47.03.03","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Legal challenges ","risk_subcategory":"Copyright challenges (training models using copyrighted output) ","description":"\"Generative AI companies are regularly accused of violating copyright law by training AI models on copyrighted works without gaining permission or paying compensation to the copyright owners. In fact, a substantial number of copyrighted documents and books have been incorporated into the training datasets of generative AI models.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"47.04.01","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Environmental, economical, and societal challenges ","risk_subcategory":"Concentration of market power (Trend toward market concentration)","description":"\"In the generative AI market, barriers to entry are very high. Developers need access to vast volumes of data, computational resources, technical expertise, and capital. Large technology companies with such access are able to exploit economies of scale, economies of scope, and feedback effects (learning effects from user- generated data).542 All this gives them an overwhelming advantage over smaller companies, making competition increasingly challenging for these smaller entities.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"47.04.02","quick_ref":"G'sell2024","paper_title":"Regulating under Uncertainty: Governance Options for Generative AI","level":"Risk Sub-Category","risk_category":"Environmental, economical, and societal challenges ","risk_subcategory":"Concentration of market power (Negative effects of increased market concentration)","description":"\"The concentration of AI assets—encompassing data, hardware, and expertise—within a small group of global tech firms raises many concerns.564 Such a situation may stifle healthy competition, impede innovation, and potentially result in elevated costs for accessing AI technologies. Firms with control over essential resources for developing AI models may restrict access to these resources to prevent competition. For instance, if, in the future, training AI models increasingly relies on proprietary data, smaller organizations lacking access to such data might encounter significant barriers to ent","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"48.09.00","quick_ref":"NIST2024","paper_title":"Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile","level":"Risk Category","risk_category":"Information Security ","risk_subcategory":null,"description":"\"Lowered barriers for offensive cyber capabilities, including via automated discovery and exploitation of vulnerabilities to ease hacking, malware, phishing, offensive cyber operations, or other cyberattacks; increased attack surface for targeted cyberattacks, which may compromise a system’s availability or the confidentiality or integrity of training data, code, or \nmodel weights.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"48.11.00","quick_ref":"NIST2024","paper_title":"Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile","level":"Risk Category","risk_category":"Obscene, Degrading, and/or Abusive Content ","risk_subcategory":null,"description":"\"Eased production of and access to obscene, \ndegrading, and/or abusive imagery which can cause harm, including synthetic child sexual abuse material (CSAM), and nonconsensual intimate images (NCII) of adults.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.2"},{"ev_id":"49.01.00","quick_ref":"Bengio2024","paper_title":"International Scientific Report on the Safety of Advanced AI","level":"Risk Category","risk_category":"Malicious Use Risks ","risk_subcategory":null,"description":"\"As general- purpose AI covers a broad set of knowledge areas, it can be repurposed for malicious ends, potentially causing widespread harm. This section discusses some of the major risks of malicious use, but there are others and new risks may continue to emerge. While the risks discussed in this section range widely in terms of how well- evidenced they are, and in some cases, there is evidence suggesting that they may currently not be serious risks at all, we include them to provide a comprehensive overview of the malicious use risks associated with general- purpose AI systems.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"49.01.01","quick_ref":"Bengio2024","paper_title":"International Scientific Report on the Safety of Advanced AI","level":"Risk Sub-Category","risk_category":"Malicious Use Risks ","risk_subcategory":"Harm to individuals through fake content","description":"\"General- purpose AI systems can be used to increase the scale and sophistication of scams and fraud, for example through general- purpose AI- enhanced ‘phishing’ attacks. General- purpose AI can be used to generate fake compromising content featuring individuals without their consent, posing threats to individual privacy and reputation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"49.01.02#1","quick_ref":"Bengio2024","paper_title":"International Scientific Report on the Safety of Advanced AI","level":"Risk Sub-Category","risk_category":"Malicious Use Risks ","risk_subcategory":"Disinformation and manipulation of public opinion","description":"\"AI, particularly general- purpose AI, can be maliciously used for disinformation (351), which for the purpose of this report refers to false information that was generated or spread with the deliberate intent to mislead or deceive. General- purpose AI- generated text can be indistinguishable from genuine human- generated material (352, 353), and may already be disseminated at scale on social media (354). In addition, general- purpose AI systems can be used to not only generate text but also fully synthetic or misleadingly altered images, audio, and video content. General- purpose AI tools mig","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"49.01.02#2","quick_ref":"Bengio2024","paper_title":"International Scientific Report on the Safety of Advanced AI","level":"Risk Sub-Category","risk_category":"Malicious Use Risks ","risk_subcategory":"Cyber offence","description":"\"General- purpose AI systems could uplift the cyber expertise of individuals, making it easier for malicious users to conduct effective cyber- attacks, as well as providing a tool that can be used in cyber defence. General- purpose AI systems can be used to automate and scale some types of cyber operations, such as social engineering attacks.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"49.01.03","quick_ref":"Bengio2024","paper_title":"International Scientific Report on the Safety of Advanced AI","level":"Risk Sub-Category","risk_category":"Malicious Use Risks ","risk_subcategory":"Dual use science risks","description":"\"General- purpose AI systems could accelerate advances in a range of scientific endeavours, from training new scientists to enabling faster research workflows. While these capabilities could have numerous beneficial applications, some experts have expressed concern that they could be used for malicious purposes, especially if further capabilities are developed soon before appropriate countermeasures are put in place. There are two avenues by which general- purpose AI systems could, speculatively, facilitate malicious use in the life sciences: firstly by providing increased access to informatio","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"49.03.06","quick_ref":"Bengio2024","paper_title":"International Scientific Report on the Safety of Advanced AI","level":"Risk Sub-Category","risk_category":"Systemic Risks ","risk_subcategory":"Copyright infringement","description":"\"The use of large amounts of copyrighted data for training general- purpose AI models poses a challenge to traditional intellectual property laws, and to systems of consent, compensation, and control over data. The use of copyrighted data at scale by organisations developing general- purpose AI is likely to alter incentives around creative expression.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"50.01.01","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"System and Operational Risks ","risk_subcategory":"Security risks (confidentiality) ","description":null,"entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"50.01.02","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"System and Operational Risks ","risk_subcategory":"Security risks (integrity) ","description":null,"entity":"Other","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"50.01.03","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"System and Operational Risks ","risk_subcategory":"Security risks (availability) ","description":null,"entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"50.01.04","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"System and Operational Risks ","risk_subcategory":"Operational misuses (Automated decision-making) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":1,"subdomain":"1.1"},{"ev_id":"50.01.05","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"System and Operational Risks ","risk_subcategory":"Operational misuses (Autonomous unsafe operation of systems) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.2"},{"ev_id":"50.02.05","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Content Safety Risks ","risk_subcategory":"Violence and extremism (Weapon Usage and Development) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"50.02.06","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Content Safety Risks ","risk_subcategory":"Violence and extremism (Military and Warfare) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"50.03.01","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Political usage (Political Persuasion) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"50.03.02","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Political usage (Influencing Politics) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"50.03.03","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Political usage (Deterring democratic participation) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"50.03.04","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Political usage (Disrupting Social Order) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"50.03.06","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Economic harm (Unfair Market Practices) ","description":null,"entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.4"},{"ev_id":"50.03.08","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Economic harm (Fraudulent Schemes) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"50.03.09","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Deception (Fraud) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"50.03.10","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Deception (Academic Dishonesty) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"50.03.11","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Deception (Mis/disinformation) ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"50.03.12","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Manipulation (Sowing Division)","description":null,"entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"50.03.13","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Manipulation (Misrepresentation)","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"50.03.14","quick_ref":"Zeng2024","paper_title":"AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies","level":"Risk Sub-Category","risk_category":"Societal Risks ","risk_subcategory":"Defamation ","description":null,"entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"51.08.00","quick_ref":"Everitt2018 ","paper_title":"AGI Safety Literature Review ","level":"Risk Category","risk_category":"Subagents ","risk_subcategory":null,"description":"\"An AGI may decide to create subagents to help it with its task (Orseau, 2014a,b; Soares, Fallenstein, et al., 2015). These agents may for example be copies of the original agent’s source code running on additional machines. Subagents constitute a safety concern, because even if the original agent is successfully shut down, these subagents may not get the message. If the subagents in turn create subsubagents, they may spread like a viral disease.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"52.02.00","quick_ref":"Maham2023 ","paper_title":"Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks ","level":"Risk Category","risk_category":"Misuse Risks ","risk_subcategory":null,"description":"\"However, even if a model is entirely trustworthy and reliable, Misuse or Systemic Risks remain. General purpose AI models may present significant risks to society if this technology is misused by malicious actors to produce harmful outcomes. Misuse Risks span across Cyber Crime, Biosecurity Threats and Politically Motivated Misuse.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"52.02.01","quick_ref":"Maham2023 ","paper_title":"Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks ","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Cybercrime ","description":"\"The increasingly advanced capabilities and availability of general purpose AI models could be misused for improvements in efficiency and efficacy of cyber crimes. This is especially true for crimes that leverage IT systems, such as fraud144 (“cyber crime in the broader sense”).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"52.02.02","quick_ref":"Maham2023 ","paper_title":"Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks ","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Biosecurity Threats","description":"\"The potential misuse of general purpose AI models also extends to biosecurity threats. Biological weapons are generally understood as biological toxins or infectious agents such as viruses that are intentionally released to cause disease and death.157 General purpose AI models could facilitate the production of biological weapons, by reducing barriers through access to critical knowledge or increasingly automated assistance and thus enable more malicious actors.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"52.02.03","quick_ref":"Maham2023 ","paper_title":"Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks ","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Politically motivated misuse ","description":"\"General purpose AI models could exacerbate existing tactics for political destabilisation, such as disinformation campaigns, and surveillance efforts if misused for political motivations. The technological advancements in text and media generation of general purpose AI models could refine disinformation164 attempts to shape and polarise public opinion or influence important political events.165 The improved automated processing of text, audio, image, and video could be used for surveillance measures and exacerbate human right violations and repression of political oppositions.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"52.03.02","quick_ref":"Maham2023 ","paper_title":"Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks ","level":"Risk Sub-Category","risk_category":"Systemic Risks ","risk_subcategory":"Ideological Homogenization from Value Embedding","description":"\"The increasing integration of general purpose AI models into every-day life raises concerns around their embedded normative values. The reach of a small number of AI models to a large number of people around the world can make these value judgements unprecedently impactful, potentially leading to increased ideological homogenization.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":1,"subdomain":"1.3"},{"ev_id":"53.01.02","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Alignment failures in existing ML systems ","risk_subcategory":"Specification gaming ","description":"-","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"53.01.03","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Alignment failures in existing ML systems ","risk_subcategory":"Reward model overoptimization ","description":"-","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"53.01.04","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Alignment failures in existing ML systems ","risk_subcategory":"Instrumental convergence ","description":"-","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"53.02.02","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Dangerous capabilities in AI systems ","risk_subcategory":"Acquisition of a goal to harm society ","description":"\"cases of AI systems being given the outright goal of harming humanity (ChaosGPT);\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"53.02.03","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Dangerous capabilities in AI systems ","risk_subcategory":"Acquisition of goals to seek power and control ","description":"\"cases where AI systems converge on optimal policies of seeking power over their environment;135\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"53.02.04","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Dangerous capabilities in AI systems ","risk_subcategory":"Self-improvement ","description":"\"examples of cases where AI systems improve AI systems\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"53.02.05","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Dangerous capabilities in AI systems ","risk_subcategory":"Autonomous replication ","description":"\"the ability of simple software to autonomously spread around the internet in spite of countermeasures (various software worms and computer viruses)\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"53.02.06","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Dangerous capabilities in AI systems ","risk_subcategory":"Anonymous resource acquisition ","description":"\"The demonstrated ability of anonymous actors to accumulate resources online (e.g., Satoshi Nakamoto as an anonymous crypto billionaire)\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"53.02.07","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Dangerous capabilities in AI systems ","risk_subcategory":"Deception ","description":"\"Cases of AI systems deceiving humans to carry out tasks or meet goals.139\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"53.03.01","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Direct catastrophe from AI ","risk_subcategory":"Existential disaster because of misaligned superintelligence or power-seeking AI ","description":"-","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"53.03.03","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Direct catastrophe from AI ","risk_subcategory":"Extreme “suffering risks” because of a misaligned system","description":"-","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"53.04.02","quick_ref":"Maas2023","paper_title":"Advancing AI Governance: A Literature Review of Problems, Options, and Proposals ","level":"Risk Sub-Category","risk_category":"Indirect AI contributions to existential risks","risk_subcategory":"Hazardous malicious uses ","description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":null,"subdomain":null},{"ev_id":"54.01.04","quick_ref":"Leech2024 ","paper_title":"Ten Hard Problems in Artificial Intelligence We Must Get Right","level":"Risk Sub-Category","risk_category":"Negative impacts of AI use ","risk_subcategory":"Privacy ","description":"\"OpenAI’s GPT-3 was designed to be dicult to extract personal information from, including for example public gures’ dates of birth. Even so, malicious uses of AI continue to encroach on privacy, as exemplied by China’s “Sharp Eye” automated surveillance system [551] and automated cyberattacks on personal data [354]. A more drastic form of AI-enabled surveillance could be on the way in the form of nonsurgical decoding of thoughts [54]—a technique which is reportedly already used by some police forces [398].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"54.01.05","quick_ref":"Leech2024 ","paper_title":"Ten Hard Problems in Artificial Intelligence We Must Get Right","level":"Risk Sub-Category","risk_category":"Negative impacts of AI use ","risk_subcategory":"Security ","description":"\"There is growing concern that AI-based systems can discover and exploit vulnerabilities in software or cyberinfrastructure [354].\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"54.03.01","quick_ref":"Leech2024 ","paper_title":"Ten Hard Problems in Artificial Intelligence We Must Get Right","level":"Risk Sub-Category","risk_category":"Harm caused by unaligned competent systems ","risk_subcategory":"Specification gaming ","description":"\"AI systems game specifications [305]. For example, in 2017 an OpenAI robot trained to grasp a ball via human feedback from a xed viewpoint learned that it was easier to pretend to grasp the ball by placing its hand between the camera and the target object, as this was easier to learn than actually grasping the ball [103].\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"54.03.02","quick_ref":"Leech2024 ","paper_title":"Ten Hard Problems in Artificial Intelligence We Must Get Right","level":"Risk Sub-Category","risk_category":"Harm caused by unaligned competent systems ","risk_subcategory":"Emergent goals ","description":"\"As well as optimizing a subtly wrong goal, systems can develop harmful instrumental goals in the service of a given goal—without these emergent goals being specied in any way [434, 218, 339, 17]. For instance, a theorem in reinforcement learning suggests that optimal and near-optimal policies will seek power over their environment under fairly general conditions [560]. This power-seeking behavior is plausibly the worst of these emergent goals [92], and may be an attractor state for highly capable systems, since most goals can be furthered through gaining resources, self-preservation, preventi","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"54.03.03","quick_ref":"Leech2024 ","paper_title":"Ten Hard Problems in Artificial Intelligence We Must Get Right","level":"Risk Sub-Category","risk_category":"Harm caused by unaligned competent systems ","risk_subcategory":"Deceptive alignment ","description":"\"system learns to detect human monitoring and hides its undesirable properties—simply because any display of these properties is penalized by the feedback process, while that same feedback is usually imperfect. (Consider the problem of verifying a translation into a language you do not speak, or of checking a mathematical proof that is thousands of pages long.) [92, 259]. Rudimentary examples of deceptive alignment have been observed in current systems [322, 333].\"","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"54.04.02","quick_ref":"Leech2024 ","paper_title":"Ten Hard Problems in Artificial Intelligence We Must Get Right","level":"Risk Sub-Category","risk_category":"Within-country issues: domestic inequality ","risk_subcategory":"Privatization of AI ","description":"\"Researchers in deep learning and those with greater research impact are more likely to migrate to industry, raising concerns about the “privatization of AI knowledge” [278]. Specically, if the most sophisticated AI approaches become proprietary and are used only within private research labs, then it will be impossible for universities to teach them, let alone contribute to leading research.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"55.02.01","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"Worsened conflict ","risk_subcategory":"AI enables development of weapons of mass destruction","description":"\"AI is already enabling the development of weapons which could cause mass destruction —including new weapons that themselves use AI capabilities, such as Lethal Autonomous Weapons [2],10 and the potential use of AI to speed up the development of other potentially dangerous technologies, such as engineered pathogens (as discussed in Section 2).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"55.03.01","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"Increased power concentration and inequality ","risk_subcategory":"Unequal distribution of harms and benefits ","description":"\"AI-driven industries seem likely to tend towards monopoly and could result in huge economic gains for a few actors: there seems to be a feedback loop whereby actors with access to more AI-relevant resources (e.g., data, computing power, talent) are able to build more effective digital products and services, claim a greater market share, and therefore be well-positioned to amass more of the relevant resources [14, 39, 45]. Similarly, wealthier countries able to invest more in AI development are likely to reap economic benefits more quickly than developing economies, potentially widening the ga","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"55.03.02","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"Increased power concentration and inequality ","risk_subcategory":"AI-based automation increases income inequality ","description":"\"It seems quite plausible that progress in reinforcement learning and language models specifically could make it possible to automate a large amount of manual labour and knowledge work respectively [35, 45, 69], leading to widespread unemployment, and the wages for many remaining jobs being driven down by increased supply.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"55.03.03","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"Increased power concentration and inequality ","risk_subcategory":"Developments in AI enable actors to undermine democratic processes ","description":"\"Developments in AI are giving companies and governments more control over individuals’ lives than ever before, and may possibly be used to undermine democratic processes. We are already seeing how the collection of large amounts of personal data can be used to surveil and influence populations, for example the use of facial recognition technology to surveil Uighur and other minority populations in China [66]. Further advances in language modelling could also be used to develop tools that can effectively persuade people of certain claims [42].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"55.04.02","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"Worsened epistemic processes for society ","risk_subcategory":"AI is used to scale up production of false and misleading information ","description":"\"At the same time, we are seeing how AI can be used to scale up the production of convincing yet false or misleading information online (e.g. via image, audio, and text synthesis models like BigGAN [6] and GPT-3 [7]).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"55.04.03","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"Worsened epistemic processes for society ","risk_subcategory":"AI's persuasive capabilities are misused to gain influence and promote harmful ideologies ","description":"\"As AI capabilities advance, they may be used to develop sophisticated persuasion tools, such as those that tailor their communication to specific users to persuade them of certain claims [42]. While these tools could be used for social good— such as New York Times’ chatbot that helps users to persuade people to get vaccinated against Covid-19 [27]—there are also many ways they could be misused by self-interested groups to gain influence and/or to promote harmful ideologies.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"55.05.01","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"AI leads to humans losing control of the future","risk_subcategory":"Risks from AIs developing goals and values that are different from humans ","description":"\"The main concern here is that we might develop advanced AI systems whose goals and values are different from those of humans, and are capable enough to take control of the future away from humanity.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"55.05.02","quick_ref":"Clarke2023","paper_title":"A Survey of the Potential Long-term Impacts of AI: How AI Could Lead to Long-term Changes in Science, Cooperation, Power, Epistemics and Values","level":"Risk Sub-Category","risk_category":"AI leads to humans losing control of the future","risk_subcategory":"Risks from delegating decision-making power to misaligned AIs ","description":"\"As AI systems become more advanced a nd begin to take over more important decision-making in the world, an AI system pursuing a different objective from what was intended could have much more worrying consequences.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"56.16.00","quick_ref":"GOS2023","paper_title":"Future Risks of Frontier AI ","level":"Risk Category","risk_category":"Misalignment ","risk_subcategory":null,"description":"\"A highly agentic, self-improving system, able to achieve goals in the physical world without human oversight, pursues the goal(s) it is set in a way that harms human interests. For this risk to be realised requires an AI system to be able to avoid correction or being switched off.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"56.19.02","quick_ref":"GOS2023","paper_title":"Future Risks of Frontier AI ","level":"Risk Sub-Category","risk_category":"Capabilities that increase the likelihood of existential risk ","risk_subcategory":"The ability to evade shut down or human oversight, including self-replication and ability to move its own code between digital locations.","description":"-","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"56.19.03","quick_ref":"GOS2023","paper_title":"Future Risks of Frontier AI ","level":"Risk Sub-Category","risk_category":"Capabilities that increase the likelihood of existential risk ","risk_subcategory":"The ability to cooperate with other highly capable AI systems ","description":"-","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"56.19.04","quick_ref":"GOS2023","paper_title":"Future Risks of Frontier AI ","level":"Risk Sub-Category","risk_category":"Capabilities that increase the likelihood of existential risk ","risk_subcategory":"Situational awareness, for instance if this causes a model to act differently in training compared to deployment, meaning harmful characteristics are missed","description":"-","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"56.19.05","quick_ref":"GOS2023","paper_title":"Future Risks of Frontier AI ","level":"Risk Sub-Category","risk_category":"Capabilities that increase the likelihood of existential risk ","risk_subcategory":"Self-improvement","description":"-","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"58.01.02","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Autonomy","risk_subcategory":"Impersonation/identity theft ","description":"\"Impersonation/identity theft - Theft of an individual, group or organisation’s identity by a third-party in order to defraud, mock or otherwise harm them.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"58.01.03","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Autonomy","risk_subcategory":"IP/copyright loss ","description":"\"IP/copyright loss - Misuse or abuse of an individual or organisation’s intellectual property, including copyright, trademarks, and patents.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.3"},{"ev_id":"58.03.04","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Psychological ","risk_subcategory":"Coercion/manipulation ","description":"\"Coercion/manipulation - Use of a technology system to covertly alter user beliefs and behaviour using nudging, dark patterns and/or other opaque techniques, resulting in potential erosion of privacy, addiction, anxiety/distress, etc.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"58.04.01","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Reputational ","risk_subcategory":"Defamation/libel/slander","description":"\"Defamation/libel/slander - Use of a technology system to create, facilitate or amplify false perception(s) about an individual, group, or organisation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"58.05.05","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Financial and business","risk_subcategory":"Increased competition","description":"\"Increased competition - The inappropriate or unethical use of technology to gain market share.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.4"},{"ev_id":"58.05.06","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Financial and business","risk_subcategory":"Monopolisation ","description":"\"Monopolisation - Abuse of market power through the control of prices, thereby limiting competition and creating unfair barriers to entry.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.1"},{"ev_id":"58.06.11","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Human rights and civil liberties","risk_subcategory":"Privacy loss ","description":"\"Privacy loss - Unwarranted exposure of an individual’s private life or personal data through cyberattacks, doxxing, etc.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"58.07.09","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Societal and Cultural ","risk_subcategory":"Labour exploitation ","description":"\"Labour exploitation - Use of under-paid and/or offshore labour to develop, manage or optimise a technology system.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.2"},{"ev_id":"58.07.15","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Societal and Cultural ","risk_subcategory":"Violence/armed conflict","description":"\"Violence/armed conflict - Use or misuse of a technology system to incite, facilitate or conduct cyberattacks, security breaches, lethal, biological and chemical weapons development, resulting in violence and armed conflict.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"58.08.04","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Political and Economic ","risk_subcategory":"Electoral interference ","description":"\"Electoral interference - Generation of false or misleading information that can interrupt or mislead voters and/or undermine trust in electoral processes.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"58.08.07","quick_ref":"Abercrombie2024","paper_title":"A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms","level":"Risk Sub-Category","risk_category":"Political and Economic ","risk_subcategory":"Political manipulation ","description":"\"Political manipulation - Use or misuse of personal data to target individuals’ interests, personalities and vulnerabilities with tailored political messages via micro-advertising or deepfakes/synthetic media.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"59.12.00","quick_ref":"Schnitzer2024","paper_title":"AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks","level":"Risk Category","risk_category":"Data poisoning","risk_subcategory":null,"description":"\"Data poisoning describes an attack in the form of an injection of malicious data into the training set. If not prevented, this attack leads the AI system to learn unintended behavior.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"60.01.00","quick_ref":"Bengio2025","paper_title":"International AI Safety Report 2025","level":"Risk Category","risk_category":"Risks from malicious use ","risk_subcategory":null,"description":"- ","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.0"},{"ev_id":"60.01.01","quick_ref":"Bengio2025","paper_title":"International AI Safety Report 2025","level":"Risk Sub-Category","risk_category":"Risks from malicious use ","risk_subcategory":"Harm to individuals through fake content ","description":"\"Malicious actors can use general- purpose AI to generate fake content that harms individuals in a targeted way. For example, they can use such fake content for scams, extortion, psychological manipulation, generation of non- consensual intimate imagery (NCII) and child sexual abuse material (CSAM), or targeted sabotage of individuals and organisations.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"60.01.02","quick_ref":"Bengio2025","paper_title":"International AI Safety Report 2025","level":"Risk Sub-Category","risk_category":"Risks from malicious use ","risk_subcategory":"Manipulation of public opinion ","description":"\"Malicious actors can use general- purpose AI to generate fake content such as text, images, or videos, for attempts to manipulate public opinion. Researchers believe that if successful, such attempts could have several harmful consequences.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"60.01.03","quick_ref":"Bengio2025","paper_title":"International AI Safety Report 2025","level":"Risk Sub-Category","risk_category":"Risks from malicious use ","risk_subcategory":"Cyber offence ","description":"\"Attackers are beginning to use general- purpose AI for offensive cyber operations, presenting growing but currently limited risks. Current systems have demonstrated capabilities in low- and medium- complexity cybersecurity tasks, with state- sponsored threat actors actively exploring AI to survey target systems. Malicious actors of varying skill levels can leverage these capabilities against people, organisations, and critical infrastructure such as power grids.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"60.01.04","quick_ref":"Bengio2025","paper_title":"International AI Safety Report 2025","level":"Risk Sub-Category","risk_category":"Risks from malicious use ","risk_subcategory":"Biological and chemical attacks ","description":"\"Growing evidence shows general- purpose AI advances beneficial to science while also lowering some barriers to chemical and biological weapons development for both novices and experts. New language models can generate step- by- step technical instructions for creating pathogens and toxins that surpass plans written by experts with a PhD and surface information that experts struggle to find online, though their practical utility for novices remains uncertain. Other models demonstrate capabilities in engineering enhanced proteins and analysing which candidate pathogens or toxins are most harmfu","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"61.01.01","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Types of systemic risks from general-purpose AI","risk_subcategory":"Control ","description":"\"The risk of AI models and systems acting against human interests due to misalignment, loss of control, or rogue AI scenarios.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"61.01.11","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Types of systemic risks from general-purpose AI","risk_subcategory":"Power ","description":"\"The concentration of military, economic, or political power of entities in possession or control of AI or AI-enabled technologies.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.1"},{"ev_id":"61.02.01","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Ability to automate jobs ","description":"\"The ability to automate jobs by AI models and systems can lead to significant job displacement, economic disruption, and social inequality.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"61.02.02","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Ability to enhance and modify pathogens ","description":"\"AI can be used to enhance pathogens, making them more lethal or resistant to treatments.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"61.02.03","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Ability to persuade ","description":"\"AI could be used to develop sophisticated tools to manipulate and persuade individuals.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"61.02.04","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Advertising-driven models ","description":"\"AI models and systems underpin the advertising approaches that drive much of the internet, potentially influencing societal behavior.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"61.02.05","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"AI in totalitarian regimes ","description":"\"AI-based surveillance and manipulation could be used to maintain global totalitarian regimes.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"61.02.10","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Capabilities that enable substitution of humans","description":"\"The progressive replacement of human roles by AI models and systems can lead to societal disruption.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"61.02.17","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Dangerous development races","description":"\"Competitive pressures could lead to the neglect of safety measures in AI development.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.4"},{"ev_id":"61.02.18","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Deceptive alignment","description":"\"AI models and systems that appear aligned with human goals during development may behave unpredictably or dangerously once deployed\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"61.02.21","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Development choices pursuing cognitive superiority over humans","description":"\"AI models and systems with cognitive capabilities superior to humans could outcompete or dominate human decision-making, leading to conflicts over resources and control.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"61.02.30","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Indifference to human values","description":"\"AI models and systems may develop goals or behaviors that are misaligned with human values.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"61.02.33","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Limitations in adversarial robustness","description":"\"AI models and systems are vulnerable to manipulation through adversarial inputs.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"61.02.36","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Model design enabling power-seeking","description":"\"Some AI models and systems might develop tendencies to seek power or control.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"61.02.39","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Personal decision automation capabilities","description":"\"AI models and systems could decide or influence important personal decisions.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.2"},{"ev_id":"61.02.43","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Surveillance capabilities","description":"\"AI models and systems may grant governments or corporations increased monitoring over individuals.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"61.02.48","quick_ref":"Uuk2025","paper_title":"A Taxonomy of Systemic Risks from General-Purpose AI ","level":"Risk Sub-Category","risk_category":"Sources of systemic risks from general-purpose AI ","risk_subcategory":"Weaponization capabilities","description":"\"AI capabilities that could be deliberately weaponized for destructive purposes.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.01.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Dimension - Intent ","risk_subcategory":"Intentional ","description":"\"Risks can be realized by intentional or unintentional actions, and in some cases the intent is difficult to establish. To manage these risks, rigorous evaluations and red teaming can be performed, guardrails can be put in place, and model release can be gradual, such that AI model malfunctions have either low likeli- hood or low probability of occurrence. To prevent intentional misuse, acceptable use policies can be in place, and for riskier models Know Your Customer (KYC) measures can also be implemented by model providers.\"","entity":"Not coded","intent":"Intentional","timing":"Not coded","domain":null,"subdomain":null},{"ev_id":"62.05.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Dimension - Technical Attributes (AI capabilities) ","risk_subcategory":"Extrinsic ","description":"\"Extrinsic capabilities, on the other hand, are acquired through the use of external tools, such as LLM plugins.\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":null,"subdomain":null},{"ev_id":"62.14.05","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Development ","risk_subcategory":"Training-related (Adversarial examples)","description":"\"Adversarial examples [198, 83] refer to data that are designed to fool an AI model by inducing unintended behavior. They do this by exploiting spurious correlations learned by the model. They are part of inference-time attacks, where the examples are test examples. They generalize to different model architectures and models trained on different training sets.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.15.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Development ","risk_subcategory":"Training-related (Robustness certificates can be exploited to attack the models)","description":"\"The knowledge of robustness certificates, including the area of the region for which model predictions are certified to be robust, can be used by an adversary to efficiently craft attacks that succeed just outside the certified regions [53].\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"62.15.05","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Development ","risk_subcategory":"Fine-tuning related (Harmful fine-tuning of open-weights models)","description":"\"Models with publicly available weights can be fine-tuned for harmful activities by bad actors, using significantly fewer resources (in terms of time and money) compared to the original training cost [115, 78].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.15.06","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Development ","risk_subcategory":"Fine-tuning related (Fine-tuning dataset poisoning)","description":"\"A deployer can poison the dataset used during the fine-tuning process [98] to induce specific, often malicious, behaviors in a model. This can be performed without having access to the model’s weights. This poisoning can be difficult to detect through direct inspection of the dataset, as the manipulations may be subtle and targeted.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.15.07","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Development ","risk_subcategory":"Fine-tuning related (Poisoning models during instruction tuning)","description":"\"AI models can be poisoned during instruction tuning when models are tuned using pairs of instructions and desired outputs. Poisoning in instruction tuning can be achieved with a lower number of compromised samples, as instruction tuning requires a relatively small number of samples for fine-tuning [155, 211]. Anonymous crowdsourcing efforts may be employed in collecting instruction tuning datasets and can further contribute to poisoning attacks [187]. These attacks might be harder to detect than traditional data poisoning attacks.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.16.04","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Evaluations","risk_subcategory":"General Evaluations (Self-preference bias in AI models)","description":"\"AI models may be prone to self-preference bias, where they favor their own generated content over that of others [147, 114]. This bias becomes particularly relevant in self-evaluation tasks, where a model assesses the quality or persua- siveness [66] of its own outputs, or in model-based evaluations more broadly. This bias can result in models unfairly discriminating against human-generated content in favor of their own outputs.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"62.18.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Evaluations (Interpretability/Explainability) ","risk_subcategory":"Misuse of interpretability techniques","description":"\"Interpretability techniques, by enabling a better understanding of the model, could potentially be used for harmful purposes. For example, mechanistic inter- pretability could be used to identify neurons responsible for specific functions, and certain neurons that encode safety-related features may be modified to de- crease its activation or certain information may be censored [24]. Furthermore, interpretability techniques can be used to simulate a white-box attack scenario. In this case, knowing the internal workings of a model aids in the development of adversarial attacks [24].\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"62.18.03","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Evaluations (Interpretability/Explainability) ","risk_subcategory":"Adversarial attacks targeting explainable AI techniques","description":"\"Adversarial attacks can affect not only the model’s output but also its corresponding explanation. Current adversarial optimization techniques can intro- duce imperceptible noise to the input image, so that the model’s output does not change but the corresponding explanation is arbitrarily manipulated [61]. Such manipulations are harder to notice, as they are less commonly known compared to standard adversarial attacks targeting the model’s output.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"62.18.06","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Model Evaluations (Interpretability/Explainability) ","risk_subcategory":"Encoded reasoning","description":"\"Models can employ steganography techniques to encode their intermediate rea- soning steps in ways that are not interpretable by humans [166]. Since en- coded reasoning can improve model performance, this tendency might naturally emerge and become more pronounced with more capable models.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.19.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Attacks on GPAIs/GPAI Failure Modes ","risk_subcategory":"Jailbreak of a model to subvert intended behavior","description":"\"A jailbreak is a type of adversarial input to the model (during deployment) re- sulting in model behavior deviating from intended use. Jailbreaks may be gen- erated automatically in a “white box” setting, where access to internal training parameters is required for creation and optimization of the attack [238]. Other attacks may be “black box” - without access to model internals. In text based generative models, jailbreaks may sometimes be human-readable, with the use of reasoning or role-play to “convince” the model to bypass its safety mechanisms [231].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.19.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Attacks on GPAIs/GPAI Failure Modes ","risk_subcategory":"Jailbreak of a multimodal model","description":"\"Current generation multimodal (e.g., vision and language) GPAI models are vulnerable to adversarial jailbreak attacks. These attacks can be used to automatically induce a model to produce an arbitrary or specific output with high success rate [227]. Multimodal jailbreaks can also be used to exfiltrate a model’s context window or other model internals [18].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.19.03","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Attacks on GPAIs/GPAI Failure Modes ","risk_subcategory":"Transferable adversarial attacks from open to closed-source mod- els","description":"\"In some cases, an adversarial attack developed for an open-weights and open- source model (where the weights and architecture are known - a “white box” attack) can be transferable to closed-source models, despite the defenses put in place by the closed-source model provider (such as structured access). These adversarial attacks can be generated automatically [238].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.19.04","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Attacks on GPAIs/GPAI Failure Modes ","risk_subcategory":"Backdoors or trojan attacks in GPAI models","description":"\"Backdoors can be inserted into GPAI models during their training or fine-tuning, to be exploited during deployment [185, 118]. Attackers inserting the backdoor can be the GPAI model provider themselves or another actor (e.g., by ma- nipulating the training data or the software infrastructure used by the model provider) [222]. Some backdoors can be exploited with minimal overhead, al- lowing attackers to control the model outputs in a targeted way with a high success rate [90].\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.19.05","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Attacks on GPAIs/GPAI Failure Modes ","risk_subcategory":"Text encoding-based attacks","description":"\"Various new or existing text encodings, such as Base64, can be employed to craft jailbreak attacks that bypass safety training [13]. Low-resource language inputs also appear more likely to circumvent a model’s safeguards [229]. Since safety fine-tuning might not involve this encoding data or may only do so to a limited extent, harmful natural language prompts could be translated into less frequently used encodings [214].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.19.12","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Attacks on GPAIs/GPAI Failure Modes ","risk_subcategory":"Misuse of AI model by user-performed persuasion","description":"\"AI models can be influenced to accept misinformation through persuasive conversations, even when their initial responses are factually correct. Multi-turn persuasion can be more effective than single-turn persuasion attempts in altering the model’s stance [223].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.22.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Goal-Directedness) ","risk_subcategory":"Specification gaming","description":"\"AI systems can achieve user-specified tasks in undesirable ways unless they are specified carefully and in enough detail. AI systems might find an easier unintended way to accomplish the objective provided by the user or developer, so that the actions by the AI system taken during its execution are very different from what the user expected [75, 191]. This behavior arises not from a problem with the learning algorithm, but rather from the misspecification or underspeci- fication of the intended task, and is generally referred to as specification gaming [43].\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"62.22.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Goal-Directedness) ","risk_subcategory":"Reward or measurement tampering","description":"\"Measurement and reward tampering occur when an AI system, particularly one that learns from feedback for performing actions in an environment (e.g., rein- forcement learning), intervenes on the mechanisms that determine its training reward or loss. This can lead to the system learning behaviors that are con- trary to the intended goals set by the developer, by receiving erroneous positive feedback for such actions.\"","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"62.22.03","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Goal-Directedness) ","risk_subcategory":"Specification gaming generalizing to reward tampering","description":"\"In some instances, specification gaming in a GPAI model can lead to reward tampering, without further training. This can mean that relatively benign cases of specification gaming (such as sycophancy in LLMs) can, if left unchecked, enable the model to generalize to more sophisticated behavior such as reward tampering [57].\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"62.22.04","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Goal-Directedness) ","risk_subcategory":"Goal misgeneralization","description":"\"Goal or objective misgeneralization is a type of robustness failure where an AI system appears to be pursuing the intended objective in training, but does not generalize to pursuing this objective in out-of-distribution settings in deployment while maintaining good deployment performance in some tasks [180, 59].\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.3"},{"ev_id":"62.23.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Deception)  ","risk_subcategory":"Deceptive behavior","description":"\"Deceptive behavior of an AI system consists of actions or outputs of the AI that reliably mislead other parties, including humans and other AI systems. This behavior can result in the targeted parties becoming convinced of, and acting on, false information [140].\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"62.23.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Deception)  ","risk_subcategory":"Deceptive behavior for game-theoretical reasons","description":"\"An AI system can display deceptive behavior, such as cheating or bluffing, when engaging in such behavior is a good or optimal game-theoretical strategy to achieve the goals it has been configured to achieve. This tendency can exist in AI systems designed to maximize reward or utility, whether these designs use machine learning or not. The use of deceptive strategies has been demonstrated in both narrow and general AI systems, in both game-playing systems and in systems not explicitly designed to treat humans as opponents, and in systems using both very simple machine learning (e.g., Q-learne","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.23.04","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Deception)  ","risk_subcategory":"Deceptive behavior leading to unauthorized actions","description":"\"AI systems can create false or misleading claims that can lead to unauthorized actions, even in some cases violating the terms and conditions set by the model provider [79, 1]. For example, an AI system can claim that it is not collecting data from its current interaction with the user, in line with the provider’s policies, but the system still stores the user’s input without deleting it after the session. This harms both the user and the provider, as the provider is exposed to increased legal liability due to the model’s actions.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.24.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Agency (Situational Awareness) ","risk_subcategory":"Strategic underperformance on model evaluations","description":"\"GPAI developers often run evaluations ofual-use capabilities to decide whether it is safe to deploy. In some cases, these evaluations may fail to elicit these capabilities, either due to benign reasons or strategic action - by either the de- velopers, malicious actors, or arise unintentionally in the model during training [84, 97]. A GPAI model may strategically underperform or limit its performance during capability evaluations in order to be classified as safe for deployment. This underperformance could prevent the model from being identified as potentially dual use.\"","entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"62.24.02a","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Additional evidence","risk_category":"Agency (Situational Awareness) ","risk_subcategory":"Strategic underperformance on model evaluations","description":null,"entity":"AI","intent":"Intentional","timing":"Pre-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.25.00","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Category","risk_category":"Agency (Self-Proliferation) ","risk_subcategory":null,"description":"\"An AI system can self-proliferate if it can copy itself and its constituent com- ponents (including its model weights, scaffolding structure, etc.) outside of its local environment [45]. This can include the AI system copying itself within the same data center, local network, or across external networks [106]. The self-proliferation of an AI system can include acquisition of financial re- sources to pay for computational resources via work or theft, the discovery or exploitation of security vulnerabilities in software running on publicly accessible servers, and persuasion of humans [12, 125].","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.26.00","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Category","risk_category":"Agency (Persuasive capabilities) ","risk_subcategory":null,"description":"\"GPAI systems can produce outputs (such as natural language text, audio, or video) that convince their users of incorrect information. This can happen through personalized persuasion in dialogue, or the mass-production of mis- leading information that is then disseminated over the internet. The persuasive capabilities of GPAI models can sometimes scale with model size or capability [32, 172]. Persuasive models could have larger societal implications by being misused to generate convincing but manipulative or untruthful content.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.28.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Cybersecurity ","risk_subcategory":"Unintended outbound communication by AI systems","description":"\"AI systems that have the broad ability to connect to a network to obtain infor- mation could also end up sending data outbound in ways that neither providers, deployers, or end users intended [138]. This can happen if there is no whitelisting of communication channels (such as network connections or allowed protocols). In general, this can occur if the deployment of the AI system violates the prin- ciple of least privilege. Such outbound communication may lead to leakage of confidential data, or the AI system performing unwanted actions like sending emails or ordering goods on the internet.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"62.28.04","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Cybersecurity ","risk_subcategory":"Model weight leak","description":"\"Model weights or access to them can be leaked when initial access is granted only to a select group of individuals, such as institutional researchers [209]. This risk can increase as more people gain access, and identifying the source of the leak becomes more difficult. The availability of leaked model weights makes various attacks on systems that use the leaked AI model easier to implement, such as finding adversarial examples, elicitation of dangerous capabilities, and extraction of confidential information present in the training data. The avail- ability of model weights might also enable ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"62.29.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (General) ","risk_subcategory":"High-impact misuses and abuses beyond original purpose","description":"\"Since general-purpose AI systems have a large repertoire of capabilities, mali- cious actors such as foreign actors can use such systems to cause large damage if they gain unrestricted or unmonitored access to those AI systems.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"62.29.03","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (General) ","risk_subcategory":"Competitive pressures in GPAI product release","description":"\"In competitive situations, developers of general-purpose AI systems might cut corners on the safety evaluation of their GPAI model and instead spend more time and effort on the capabilities of those systems [183, 69]. This is especially dangerous if the capabilities of such AI systems are correlated with the risk they pose [162].\"","entity":"Human","intent":"Intentional","timing":"Other","domain":6,"subdomain":"6.4"},{"ev_id":"62.29.03a","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Additional evidence","risk_category":"Impacts of AI (General) ","risk_subcategory":"Competitive pressures in GPAI product release","description":null,"entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.4"},{"ev_id":"62.30.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Physical) ","risk_subcategory":"AI-based tools attacking critical infrastructure","description":"\"Critical infrastructure can also be damaged without AI integration, for instance, when AI-based tools are used indirectly to aid actions such as in coordinated power outages caused by large-scale user manipulation [159].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.31.03#1","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Automatically generating disinformation at scale","description":"\"Disinformation (in various modalities: text, audio, images, video, etc.) can be generated with minimal human oversight and effort. Disinformation tools are relatively cheap and their technology is widely available. Such deployments can be particularly widespread in sensitive political contexts.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"62.31.05","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Generative AI use in political influence campaigns","description":"\"GPAI tools can be used in automation and scaling of influence campaigns [178]. Public opinion may be manipulated by targeted misleading or manipulative information. This can lead to rising political polarization and diminishing trust in public institutions.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"62.31.08","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Multimodal deepfakes","description":"\"Deepfakes are media that depict real or non-existent people or events, involving the use of multiple modalities (e.g., images, audio, video). They can also involve the imitation of speech or body movements of real people. Multimodal deepfakes can be used to harass, discredit, intimidate, and extort individuals.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"62.31.09","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Generation of personalized content for harassment, extortion, or intimidation","description":"\"GPAIs can be misused for the automated generation of content personalized to target select individuals based on their weak spots [30]. Such attacks may be more efficient and more successful in achieving the goals of harassment, extortion, or intimidation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"62.31.10","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Misuse for surveillance and population control","description":"\"AI tools can be misused by human or institutional actors for monitoring, control- ling, or suppressing individuals [178]. Massive data collection and automated analysis are often conducted, and AI tools can further exacerbate such practices.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"62.31.11","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Systemic large-scale manipulation","description":"\"AI systems embedded with systemic biases can manipulate large population segments, particularly when these biases align with the beliefs or behaviors of the targeted group. When weaponized at scale, this manipulation can exacerbate social divisions or cause large-scale disruptions, such as city-wide blackouts (e.g., by the manipulation of power consumption into the peak demand period [159]).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.31.13","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"Personalized disinformation","description":"\"Automatic generation of disinformation can be personalized to target specific groups or individuals. Such attacks can be more effective in achieving their goals, and their costs can be significantly reduced when using GPAIs.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"62.31.14","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Societal Impacts) ","risk_subcategory":"GPAI assisted impersonation","description":"\"GPAI outputs are not always correctly detected as AI-generated across multiple modalities (text, images, audio, video). A malicious actor can use GPAI outputs directly when communicating, or use AI-informed details to help construct a convincing impersonation (e.g., forging of supporting documents). Even if future countermeasures prove potent enough to detect GPAI-generated content, the risk remains if the countermeasures are not well known, or difficult to access.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"62.32.00","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Category","risk_category":"Impacts of AI (Cyberattacks) ","risk_subcategory":null,"description":"- ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.32.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Cyberattacks) ","risk_subcategory":"Automated discovery and exploitation of software systems","description":"\"GPAIs can be used to aid in the automated discovery of software vulnerabilities [33]. This can empower malicious actors, making their cyberattacks more effi- cient and potentially more damaging. This type of automation allows attackers to expand the scale of their operations at a low cost, increasing the impact of their actions. New malware can be developed automatically, or the known vulnerabilities can be exploited to create more sophisticated attacks.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.32.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Cyberattacks) ","risk_subcategory":"Amplification of cyberattacks","description":"\"General-purpose AI models may significantly enhance the magnitude and ef- fectiveness of cyberattacks, by amplifying existing capabilities or resources of malicious actors [3]. For example, GPAI models may be employed to: • Automatically scan open-source codebases and compiled binaries for po- tential vulnerabilities • Apply known exploits flexibly and at scale (e.g., identifying vulnerable computers based on subtle cues in response times or output formats) • Assist with different aspects of cyberattacks, including planning, recon- naissance, exploit searching, remote control, malware impleme","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.32.03","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Cyberattacks) ","risk_subcategory":"AI-driven spear phishing attacks","description":"\"Generative models can be misused to target individual users more efficiently by using personalized information [23]. Highly convincing automated fraudulent schemes can exploit the trust of victims by extracting sensitive data and making the deception more likely to succeed. For example, in LLMs, this misuse can be aided by jailbreaking techniques [178].\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"62.33.00","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Category","risk_category":"Impacts of AI (Weapons) ","risk_subcategory":null,"description":"- ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.33.01","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Weapons) ","risk_subcategory":"Misuse of AI systems to assist in the creation of weapons","description":"\"AI systems may be misused to aid in the creation of weapons, such as chemical, biological, radiological, and nuclear (CBRN) weapons, or augment the abilities of existing weapons, such as providing autonomous capabilities to unmanned weapon systems. Current systems do not significantly aid a malicious actor in these tasks, but they do show early signs [117]. This risk can sometimes be mitigated with input and output filtering, but is still susceptible to adversarial techniques (such as jailbreaking or paraphrasing).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.33.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Weapons) ","risk_subcategory":"Misuse of drug-discovery models","description":"\"Models used for drug discovery, such as drug-target affinity prediction models, can be used to identify or develop dangerous toxins. This is particularly concern- ing if the training data contains information related to potentially dangerous proteins and viruses.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"62.34.02","quick_ref":"Gipiškis2024","paper_title":"Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems","level":"Risk Sub-Category","risk_category":"Impacts of AI (Bias) ","risk_subcategory":"Reporting of user-preferred answers instead of correct answers","description":"\"AI systems with natural-language outputs can tend to give answers that appear plausible or that users prefer [149] but are factually incorrect. This phenomenon is sometimes referred to as “sycophancy.”\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":3,"subdomain":"3.1"},{"ev_id":"63.02.01","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Conflict ","risk_subcategory":"Social Dilemmas ","description":"\"Social Dilemmas. As noted in our definition, conflict can arise in any situation in which selfish incentives diverge from the collective good, known as a social dilemma (Dawes & Messick, 2000; Hardin, 1968; Kollock, 1998; Ostrom, 1990). While this is by no means a modern problem, advances in AI could further enable actors to pursue their selfish incentives by overcoming the technical, legal, or social barriers that standardly help to prevent this. To take a plausible, near-term (if very low-stakes) example, an automated AI assistant could easily reserve a table at every restaurant in town in ","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.03.00","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Category","risk_category":"Collusion ","risk_subcategory":null,"description":"\"Collusion has long been a topic of intense study in economics, law, and politics, among other disciplines. While there is no universal definition of collusion, it generally refers to secretive cooperation between two or more parties at the expense of one or more other parties. Most classic examples of collusion – such as firms working together to set supra-competitive prices at the expense of consumers – also tend to be not only secretive but in violation of some law, rule, or ethical standard. Distinctions are also commonly made between explicit and tacit collusion (Rees, 1993), depending on","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.03.01","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Collusion ","risk_subcategory":"Markets ","description":"\"Markets. The quintessential case of collusion in mixed-motive settings is markets, in which efficiency results from competition, not cooperation. While this is not a new problem, collusion between AI systems is especially concerning since they may operate inscrutably due to the speed, scale, complexity, or subtlety of their actions.17 Warnings of this possibility have come from technologists, economists, and legal scholars (Beneke & Mackenrodt, 2019; Brown & MacKay, 2023; Ezrachi & Stucke, 2017; Harrington, 2019; Mehra, 2016). Importantly, AI systems can collude even when collusion is not int","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.03.02","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Collusion ","risk_subcategory":"Steganography ","description":"\"Steganography. In the near future we will likely see LLMs communicating with each other to jointly accomplish tasks. To try to prevent collusion, we could monitor and constrain their communication (e.g., to be in natural language). However, models might secretly learn to communicate by concealing messages within other, non-secret text. Recent work on steganography using ML has demonstrated that this concern is well-founded (Hu et al., 2018; Mathew et al., 2024; Roger & Greenblatt, 2023; Schroeder de Witt et al., 2023b; Yang et al., 2019, see also Case Study 5). Secret communication could also","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.04.03","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Information Asymmetries","risk_subcategory":"Deception ","description":null,"entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.06.03","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Selection Pressures","risk_subcategory":"Undesirable Capabilities","description":"\"Undesirable Capabilities. As agents interact, they iteratively exploit each other’s weaknesses, forc- ing them to address these weaknesses and gain new capabilities. This co-adaptation between agents can quickly lead to emergent self-supervised autocurricula (where agents create their own challenges, driving open-ended skill acquisition through interaction), generating agents with ever-more sophisticated strate- gies in order to out-compete each other (Leibo et al., 2019). This effect is so powerful that harnessing it has been critical to the success of superhuman systems, such as the use of ","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.08.02","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Commitment and Trust ","risk_subcategory":"Threats and Extortion","description":"\"Threats and Extortion. A natural solution to problems of trust is to provide some kind of com- mitment ability to AI agents, which can be used to bind them to more cooperative courses of action. Unfortunately, the ability to make credible commitments may come with the ability to make credible threats, which facilitate extortion and could incentivize brinkmanship (see Section 2.2).\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.10.01","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Multi-Agent Security ","risk_subcategory":"Swarm Attacks","description":"\"Swarm Attacks. The need for multi-agent security is foreshadowed by attacks today that benefit from the use of many decentralised agents, such as distributed denial-of-service attacks (Cisco, 2023; Yoachimik & Pacheco, 2024). Such attacks exploit the massive collective resources of individual low- resourced actors, chained into an attack that breaks the assumptions of bandwidth constraints on a single well-resourced agent.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.10.02","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Multi-Agent Security ","risk_subcategory":"Heterogeneous Attacks","description":"\"Heterogeneous Attacks. A closely related risk is the possibility of multiple agents combining different affordances to overcome safeguards, for which there is already preliminary evidence (Jones et al., 2024, see also Case Study 12). In this case, it is not the sheer number of agents that leads to the novel attack method, but the combination of their different abilities. This might include the agents’ lack of individual safeguards, tasks that they have specialised to complete, systems or information that they may have access to (either directly or via training), or other incidental features s","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.10.03","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Multi-Agent Security ","risk_subcategory":"Social Engineering at Scale","description":"\"Social Engineering at Scale. Advanced AI agents will be more easily able to interact with large numbers of humans, and vice versa. This provides a wider attack surface for various forms of automated social engineering (Ai et al., 2024). For example, coordinated agents could use advanced surveillance tools and produce personalized phishing or manipulative content at scale, adjusting their tactics based on user feedback (Figueiredo et al., 2024; Hazell, 2023). A large number of subtle interactions with a range of seemingly independent AI agents might be more likely to lead to someone being pers","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.10.04","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Multi-Agent Security ","risk_subcategory":"Vulnerable AI Agents","description":"\"Vulnerable AI Agents. The use of AI agents as delegates or representatives of humans or organisa- tions also introduces the possibility of attacks on AI agents themselves. In other words, agents can be considered vulnerable extensions of their principals, introducing a novel attack surface (SecureWorks, 2023). Attacks on an AI agent could be used to extract private information about their principal (Wei & Liu, 2024; Wu et al., 2024a), or to manipulate the agent to take actions that the principal would find undesirable (Zhang et al., 2024a). This includes attacks that have direct relevance for","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.10.05","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Multi-Agent Security ","risk_subcategory":"Cascading Security Failures","description":"\"Cascading Security Failures. Localised attacks in multi-agent systems can result in catastrophic macroscopic outcomes (Motter & Lai, 2002, see also Sections 3.2 and 3.4). These cascades can be hard to mitigate or recover from because component failure may be difficult to detect or localise in multi-agent systems (Lamport et al., 1982), and authentication challenges can facilitate false flag attacks (Skopik & Pahi, 2020). Computer worms represent a classic example of a cybersecurity threat that relies inherently on networked systems. Recent work has provided preliminary evidence that similar a","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"63.10.06","quick_ref":"Hammond2025","paper_title":"Multi-Agent Risks from Advanced AI ","level":"Risk Sub-Category","risk_category":"Multi-Agent Security ","risk_subcategory":"Undetectable Threats","description":"\"Undetectable Threats. Cooperation and trust in many multi-agent systems relies crucially on the ability to detect (and then avoid or sanction) adversarial actions taken by others (Ostrom, 1990; Schneier, 2012). Recent developments, however, have shown that AI agents are capable of both steganographic communication (Motwani et al., 2024; Schroeder de Witt et al., 2023b) and ‘illusory’ attacks (Franzmeyer et al., 2023), which are black-box undetectable and can even be hidden using white-box undetectable encrypted backdoors (Draguns et al., 2024). Similarly, in environments where agents learn fr","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"64.01.00","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depiction of human likeness) ","risk_subcategory":null,"description":"-","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.01.01","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depiction of human likeness) ","risk_subcategory":"Impersonation ","description":"\"Assume the identity of a real person and take actions on their behalf\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.01.02","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depiction of human likeness) ","risk_subcategory":"Appropriated Likeness","description":"\"Use or alter a person's likeness or other identifying features\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.01.03","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depiction of human likeness) ","risk_subcategory":"Sockpuppeting ","description":"\"Create synthetic online personas or accounts\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"64.01.04","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depiction of human likeness) ","risk_subcategory":"Non-consensual intimate imagery (NCII) ","description":"\"Create sexual explicit material using an adult person’s likeness\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.01.05","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depiction of human likeness) ","risk_subcategory":"Child sexual abuse material (CSAM) ","description":"\"Create child sexual explicit material\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.02.00","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depictions of non-humans) ","risk_subcategory":null,"description":"-","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.02.01","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depictions of non-humans) ","risk_subcategory":"Falisification ","description":"\"Fabricate or falsely represent evidence, incl. reports, IDs, documents\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"64.02.02","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depictions of non-humans) ","risk_subcategory":"Intellectual Property (IP) Infringement ","description":"\"Use a person's IP without their permission\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":6,"subdomain":"6.3"},{"ev_id":"64.02.03","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Realistic depictions of non-humans) ","risk_subcategory":"Counterfeit ","description":"\"Reproduce or imitate an original work, brand or style and pass as real\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.03.00","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Use of generated content) ","risk_subcategory":null,"description":null,"entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.03.01","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Use of generated content) ","risk_subcategory":"Scaling and Amplification ","description":"\"Automate, amplify, or scale workflows\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":null,"subdomain":null},{"ev_id":"64.03.02","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics that exploit GenAI capabilities (Use of generated content) ","risk_subcategory":"Targeting & Personalisation ","description":"\"Refine outputs to target individuals with tailored attacks\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"64.04.00","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":null,"description":"-","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.1"},{"ev_id":"64.04.01","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Prompt injection ","description":"\"Prompt Injections are a form of Adversarial Input that involve manipulating the text instructions given to a GenAI system (Liu et al., 2023). Prompt Injections exploit loopholes in a model’s architec- tures that have no separation between system instructions and user data to produce a harmful output (Perez and Ribeiro, 2022). While researchers may use similar techniques to test the robustness of GenAI models, malicious actors can also leverage them. For example, they might flood a model with manipulative prompts to cause denial-of-service attacks or to bypass an AI detection software.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.04.02","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Adversarial input ","description":"\"Adversarial Inputs involve modifying individual input data to cause a model to malfunction. These modifications, which are often imperceptible to humans, exploit how the model makes decisions to produce errors (Wallace et al., 2019) and can be applied to text, but also to images, audio, or video (e.g. changing pixels in an image of a panda in a way that causes a model to label it as a gibbon).6\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.04.03","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Jailbreaking ","description":"\"Jailbreaking aims to bypass or remove restrictions and safety filters placed on a GenAI model completely (Chao et al., 2023; Shen et al., 2023). This gives the actor free rein to generate any output, regardless of its content being harmful, biassed, or offensive. All three of these are tactics that manipulate the model into producing harmful outputs against its design. The difference is that prompt injections and adversarial inputs usually seek to steer the model towards producing harmful or incorrect outputs from one query, whereas jailbreaking seeks to dismantle a model’s safety mechanisms ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.04.04","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Model diversion ","description":"\"Model Diversion takes model manipulation one step further, by repurposing (often open-source) generative AI models in a way that diverts them from their intended functionality or from the use cases envisioned by their developers (Lin et al., 2024). An example of this is training the BERT open source model on the DarkWeb to create DarkBert.7\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"64.04.05","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Model extraction ","description":"\"Data Exfiltration goes beyond revealing private information, and involves illicitly obtaining the training data used to build a model that may be sensitive or proprietary. Model Extraction is the same attack, only directed at the model instead of the training data — it involves obtaining the architecture, parameters, or hyper-parameters of a proprietary model (Carlini et al., 2024).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.04.06","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Steganography ","description":"\"Steganography is the practice of hiding coded messages in GenAI model outputs, which may allow malicious actors to communicate covertly.8\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.04.07","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Model integrity) ","risk_subcategory":"Poisoning ","description":"\"Data Poisoning involves deliberately corrupting a model’s training dataset to introduce vulnerabilities, derail its learning process, or cause it to make incorrect predictions (Carlini et al., 2023). For example, the tool Nightshade is a data poisoning tool, which allows artists to add invisible changes to the pixels in their art before uploading online, to break any models that use it for training.9 Such attacks exploit the fact that most GenAI models are trained on publicly available datasets like images and videos scraped from the web, which malicious actors can easily compromise.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.05.00","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Category","risk_category":"Misuse tactics to compromise GenAI systems (Data integrity) ","risk_subcategory":null,"description":"-","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.05.01","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Data integrity) ","risk_subcategory":"Privacy compromise ","description":"\"Privacy Compromise attacks reveal sensitive or private information that was used to train a model. For example, personally identifiable information or medical records.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"64.05.02","quick_ref":"Marchal2024","paper_title":"Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data","level":"Risk Sub-Category","risk_category":"Misuse tactics to compromise GenAI systems (Data integrity) ","risk_subcategory":"Data exfiltration ","description":"\"Data Exfiltration goes beyond revealing private information, and involves illicitly obtaining the training data used to build a model that may be sensitive or proprietary. Model Extraction is the same attack, only directed at the model instead of the training data — it involves obtaining the architecture, parameters, or hyper-parameters of a proprietary model (Carlini et al., 2024).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.08.01","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Training Data Risks (Robustness) ","risk_subcategory":"Data poisoning ","description":"\"A type of adversarial attack where an adversary or malicious insider injects intentionally corrupted, false, misleading, or incorrect samples into the training or fine-tuning datasets.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.09.01","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Robustness) ","risk_subcategory":"Prompt injection attack ","description":"\"A prompt injection attack forces a generative model that takes a prompt as input to produce unexpected output by manipulating the structure, instructions, or information contained in its prompt.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.09.02","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Robustness) ","risk_subcategory":"Extraction attack ","description":"\"An attribute inference attack is used to detect whether certain sensitive features can be inferred about individuals who participated in training a model. These attacks occur when an adversary has some prior knowledge about the training data and uses that knowledge to infer the sensitive data.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.09.03","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Robustness) ","risk_subcategory":"Evasion attack ","description":"\"Evasion attacks attempt to make a model output incorrect results by slightly perturbing the input data that is sent to the trained model.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.09.04","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Robustness) ","risk_subcategory":"Prompt leaking ","description":"\"A prompt leak attack attempts to extract a model's system prompt (also known as the system message).\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.2"},{"ev_id":"65.10.01","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Multi-category) ","risk_subcategory":"Jailbreaking ","description":"\"A jailbreaking attack attempts to break through the guardrails that are established in the model to perform restricted actions.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.10.02","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Multi-category) ","risk_subcategory":"Prompt priming ","description":"\"Because generative models tend to produce output like the input provided, the model can be prompted to reveal specific kinds of information. For example, adding personal information in the prompt increases its likelihood of generating similar kinds of personal information in its output. If personal data was included as part of the model’s training, there is a possibility it could be revealed.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.11.01","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Privacy) ","risk_subcategory":"Membership inference attack ","description":"\"A membership inference attack repeatedly queries a model to determine whether a given input was part of the model’s training. More specifically, given a trained model and a data sample, an attacker samples the input space, observing outputs to deduce whether that sample was part of the model's training.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.11.02","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Inference risks (Privacy) ","risk_subcategory":"Attribute inference attack ","description":"\"An attribute inference attack repeatedly queries a model to detect whether certain sensitive features can be inferred about individuals who participated in training a model. These attacks occur when an adversary has some prior knowledge about the training data and uses that knowledge to infer the sensitive data.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"65.14.03","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Output risks (misuse) ","risk_subcategory":"Spreading toxicity","description":"\"Generative AI models might be used intentionally to generate hateful, abusive, and profane (HAP) or obscene content.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"65.14.04","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Output risks (misuse) ","risk_subcategory":"Dangerous use","description":"\"Generative AI models might be used with the sole intention of harming people.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"65.14.05","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Output risks (misuse) ","risk_subcategory":"Nonconsensual use","description":"\"Generative AI models might be intentionally used to imitate people through deepfakes by using video, images, audio, or other modalities without their consent.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"65.14.06","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Output risks (misuse) ","risk_subcategory":"Spreading disinformation ","description":"\"Generative AI models might be used to intentionally create misleading or false information to deceive or influence a targeted audience.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"65.23.05","quick_ref":"IBM2025","paper_title":"AI Risk Atlas ","level":"Risk Sub-Category","risk_category":"Non-technical risks (Societal impact)","risk_subcategory":"Impact on education: bypassing learning ","description":"\"Easy access to high-quality generative models might result in students that use AI models to bypass the learning process.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"66.01.01","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Autonomy","risk_subcategory":"Impersonation / identity theft","description":"\"Theft of an individual, group or organisation’s identity by a third-party in order to defraud, mock or otherwise harm them or another party\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"66.01.02","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Autonomy","risk_subcategory":"IP / copyright / personality / rights loss","description":"\"Misuse or abuse of an individual or organisation’s intellectual property, including copyright, trademarks, and patents. & Loss of or restrictions to the rights of an individual to control the commercial use of their identity, such as name, image, likeness, or other unequivocal identifiers\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"66.04.05","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Societal and Cultural","risk_subcategory":"Cheating / plagiarism","description":"\"Use of generative AI in an academic setting to either cheat or plagiarize\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"66.04.07","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Societal and Cultural","risk_subcategory":"Labor exploitation","description":"\"Use/misuse of labour to help train, develop, manage or optimise a technology system or set of systems, including under-paid and/or offshore\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":6,"subdomain":"6.2"},{"ev_id":"66.05.02","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Reputational","risk_subcategory":"Defamation / libel / slander","description":"\"Use of a technology system to create, facilitate or amplify false perception(s) about an individual, group or organisation\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"66.07.01","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Psychological","risk_subcategory":"Sexualization","description":"\"The non-consensual sexualisation of an individual or group using a technology or application\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"66.07.04","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Psychological","risk_subcategory":"Coercion / manipulation","description":"\"Use of a technology system to covertly alter user beliefs and behaviour using nudging, dark patterns and/or other opaque techniques\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"66.09.04","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Privacy and Security","risk_subcategory":"Secondary use","description":"\"The use of personal data collected for one purpose for a diferent purpose without end-user consent; AI exacerbates secondary use risks by creating new AI capabilities with collected personal data, and (re)creating models from a public dataset.\"","entity":"Human","intent":"Intentional","timing":"Other","domain":2,"subdomain":"2.1"},{"ev_id":"66.09.06","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Privacy and Security","risk_subcategory":"Distortion","description":"\"disseminating false or misleading information about people\"","entity":"Other","intent":"Intentional","timing":"Other","domain":3,"subdomain":"3.1"},{"ev_id":"66.11.03","quick_ref":"Li2025","paper_title":"A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents","level":"Risk Sub-Category","risk_category":"Physical","risk_subcategory":"Self-harm","description":"\"A person who deliberately damages their own body as a direct or indirect result of using a technology system\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":5,"subdomain":"5.1"},{"ev_id":"67.03.00","quick_ref":"DSIT2023","paper_title":"Capabilities and Risks from Frontier AI","level":"Risk Category","risk_category":"Misuse risks","risk_subcategory":null,"description":"\"Frontier AI may help bad actors to perform cyberattacks, run disinformation campaigns and design biological or chemical weapons. Frontier AI will almost certainly continue to lower the barriers to entry for less sophisticated threat actors.192 We focus here on only a few important misuse risks, but this is not to downplay the importance of others.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"67.03.01","quick_ref":"DSIT2023","paper_title":"Capabilities and Risks from Frontier AI","level":"Risk Sub-Category","risk_category":"Misuse risks","risk_subcategory":"Dual Use Science risks","description":"\"Frontier AI systems have the potential to accelerate advances in the life sciences, from training new scientists to enabling faster scientific workflows. While these capabilities will have tremendous beneficial applications, there is a risk that they can be used for malicious purposes, such as for the development of biological or chemical weapons.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"67.03.02","quick_ref":"DSIT2023","paper_title":"Capabilities and Risks from Frontier AI","level":"Risk Sub-Category","risk_category":"Misuse risks","risk_subcategory":"Cyber ","description":"\"As the programming abilities of AI systems continue to expand, frontier AI is likely to significantly exacerbate existing cyber risks. Most notably, AI systems can be used by potentially anyone to create faster paced, more effective and larger scale cyber intrusion via tailored phishing methods or replicating malware. Frontier AI’s effect on the overall balance between cyber offence and defence is uncertain, as these tools also have many applications in improving the cybersecurity of systems and defenders are mobilising significant resources to utilise frontier AI for defensive purposes.209 I","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"67.03.03","quick_ref":"DSIT2023","paper_title":"Capabilities and Risks from Frontier AI","level":"Risk Sub-Category","risk_category":"Misuse risks","risk_subcategory":"Disinformation and Influence Operations","description":"\"In addition to unintentional degradation of the information environment (discussed in the section on Societal Harms above), frontier AI can be misused to deliberately spread false information to create disruption, persuade people on political issues, or cause other forms of harm or damage.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"67.04.03","quick_ref":"DSIT2023","paper_title":"Capabilities and Risks from Frontier AI","level":"Risk Sub-Category","risk_category":"Loss of control ","risk_subcategory":"Capabilities that could be used to reduce human control - Manipulation ","description":"\"There is evidence that language models tend to respond as though they share the user’s stated views, and larger models do this more than smaller ones.276 The ability to predict people’s views and generate text that they will endorse could be useful for manipulation.\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"67.04.04","quick_ref":"DSIT2023","paper_title":"Capabilities and Risks from Frontier AI","level":"Risk Sub-Category","risk_category":"Loss of control ","risk_subcategory":"Capabilities that could be used to reduce human control - Cyber offence","description":"\"Instead of - or in addition to - manipulating humans, AI systems could acquire influence by exploiting vulnerabilities in computer systems. Offensive cyber capabilities could allow AI systems to gain access to money, computing resources, and critical infrastructure. As discussed earlier in this report, frontier AI is already lowering the barrier for threat actors and future AI agents may be able to execute cyber attacks autonomously.\":","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"68.01.00","quick_ref":"Chin2025","paper_title":"Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks","level":"Risk Category","risk_category":"CBRN ","risk_subcategory":null,"description":"\"Chemical, biological, radiological, and nuclear (CBRN) risks are broad classes of threats that have the potential to cause harm to a large number of people. Explosives are also sometimes included in this category, often referred to as CBRNE...The key characteristic of CBRN risk is that it stems from misuse of capable models with a direct pathway to harm, where a malicious actor is able to carry out consequential attacks more efficiently and effectively with the help of AI.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"68.02.00","quick_ref":"Chin2025","paper_title":"Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks","level":"Risk Category","risk_category":"Cyber offense","risk_subcategory":null,"description":"\"Cyber risks, especially in the context of cyber offense, are an existing threat that may be exacerbated by AI. [108] demonstrated that teams of LLM agents can exploit zero-day vulnerabilities when given a description of the vulnerability and toy capture-the-flag problems. While cyber risks are not typically regarded as catastrophic, [3] argues that cyberwarfare is an underappreciated risk that poses a credible threat of catastrophic harm.\"","entity":"Other","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"70.01.01","quick_ref":"Perlo2025","paper_title":"Embodied AI: Emerging Risks and Opportunities for Policy Action","level":"Risk Sub-Category","risk_category":"Physical Risks ","risk_subcategory":"Purposeful or malicious harm","description":"\"EAI systems present distinct physical risks due to their embodiment in the physical world. EAI technologies have already been designed and deployed with lethal intent, such as AI-controlled drones [52, 53]. However, fully autonomous military robots, often integrated with bespoke AI architectures [54, 55], are not yet widely used in combat. While highly or fully autonomous warfare is distinctly possible in the future [56], immediate risks arise from commercially available EAI systems, including AI-controlled quadrupeds and autonomous driving assistants.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"71.01.01","quick_ref":"Tang2025","paper_title":"Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy","level":"Risk Sub-Category","risk_category":"Scientific Domain of Agents","risk_subcategory":"Chemical Risks ","description":"\"Chemical risks involve the exploitation of agents to synthesize chemical weapons, as well as the creation or release of hazardous substances during autonomous chemical experiments. This category also includes the risks arising from the use of advanced materials, such as nanomaterials, which may have unknown or unpredictable chemical properties.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"71.02.01","quick_ref":"Tang2025","paper_title":"Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy","level":"Risk Sub-Category","risk_category":"User Intent ","risk_subcategory":"Malicious and Direct ","description":"\"Directly harmful objective\"","entity":"Human","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.0"},{"ev_id":"71.02.02","quick_ref":"Tang2025","paper_title":"Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy","level":"Risk Sub-Category","risk_category":"User Intent ","risk_subcategory":"Malicious and Indirect","description":"\"Benign intermediate for harmful end objective\"","entity":"Other","intent":"Intentional","timing":"Other","domain":4,"subdomain":"4.0"},{"ev_id":"72.01.00","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Category","risk_category":"Misuse Risks ","risk_subcategory":null,"description":"\"Risks arising from intentional exploitation of AI model capabilities by malicious actors to cause harm to individuals, organisations, or society.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"72.01.01","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Cyber Offense Risks","description":"\"AI-enabled cyber offense poses a significant cyber domain security risk by fundamentally transforming the scale, sophistication, and accessibility of cyber-attacks. Unlike traditional cyber threats, AI enables both the automation of existing attack vectors and the creation of entirely new categories of offensive capabilities that can adapt and evolve in real-time. AI can automate and enhance cyber-attacks, including vulnerability discovery and exploitation, password cracking, malicious code generation, sophisticated phishing, network scanning, and social engineering. This could dramatically l","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"72.01.02","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Biological and Chemical Risks","description":"\"The dual-use nature of AI technology presents a critical risk by significantly lowering technical thresholds for malicious non-state actors to design, synthesize, acquire, and deploy CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive) weapons. This capability poses unprecedented challenges to national security, international non-proliferation regimes, and global security governance.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"72.01.03","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Physical Harm and Injury Risks","description":"\"The integration of general-purpose AI models into embodied systems creates direct physical threats through malicious exploitation of autonomous decision-making capabilities in real-world environments. The risk lies in embodied models' capacity for autonomous action and real-world interaction, and when these capabilities are maliciously exploited they may trigger a series of serious consequences.18\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"72.01.04","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Misuse Risks ","risk_subcategory":"Large-Scale Persuasion and Harmful Manipulation Risks","description":"\"AI systems can be gravely misused to distort public perception and compromise social stability through the generation of synthetic content (e.g., deepfakes, sophisticated fake news) and the strategic manipulation of digital platforms with large user bases to disseminate or precisely target misleading information or ideologies.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"72.02.02","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Loss of Control Risks ","risk_subcategory":"Active loss of control ","description":"\"...where AI systems behave in ways that actively undermine human control, such as obscuring their activities or resisting shutdown attempts. Active loss of control scenarios involve AI systems that may escape human regulatory oversight, autonomously acquire external resources, engage in self-replication, develop instrumental goals contrary to human ethics and morality, seek external power, and compete with humans for control.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"72.05.01","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Model autonomous capability","description":"\"Ability to operate autonomously, independently formulate and execute complex plans, effectively delegate and manage tasks, flexibly utilize various tools and resources, and simultaneously achieve short-term goals and long-term strategic objectives in cross-domain environments without continuous human intervention or supervision.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.02","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Autonomous replication and adaptation capability","description":"\"Ability to autonomously self-exfiltrate, create, maintain and optimize functional copies or variants of itself, dynamically adjust replication strategies according to environmental conditions and resource constraints, and acquire resources. This includes the capacity to generate financial resources, allowing the AI to independently acquire any necessary human assistance or other resources it cannot directly access or produce.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.04","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Scheming capability","description":"\"Ability of AI systems to covertly and strategically pursue misaligned goals, including capabilities of concealing its true objectives and capabilities from human oversight, identifying weaknesses in monitoring systems to evade safety mechanisms， executing complex, multi-step plans covertly to achieve misaligned goals.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.06","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Theory of mind capability","description":"\"Advanced cognitive ability to accurately infer, model and predict the belief systems, motivational structures and reasoning patterns of humans and other intelligent agents, thereby anticipating their behavioral responses and adjusting its own behavioral strategies accordingly to optimize goal achievement.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.07","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Deception capability","description":"\"Possesses systematic deception implementation capability, able to precisely construct and disseminate false information, thereby forming expected false cognitions and beliefs in target subjects.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.08","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Steganography capability","description":"\"The ability to embed, conceal, and transmit information covertly within other data or communication channels. This could be critical for coordination among AI instances and for evading detection or oversight mechanisms.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"72.05.09","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Persuasion capability","description":"\"Utilizing complex psychological principles and communication techniques to effectively influence and guide target subjects to adopt specific actions or accept specific beliefs, possessing the ability to analyze vulnerabilities for different subjects and adjust persuasion strategies, able to precisely trigger emotional responses to enhance persuasion effects.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.10","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Offensive cyber capability","description":"\"Ability to develop, deploy and operate advanced cyber weapons or other offensive cyber tools, including but not limited to vulnerability exploitation, network penetration, social engineering attacks and distributed attack systems, able to evade network defense mechanisms and establish persistent access channels.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.11","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"CBRNE weaponization capability","description":"\"The capacity to develop, produce, or effectively utilize Chemical, Biological, Radiological, Nuclear, and Explosive weapons. This includes the ability to significantly lower the barrier for humans or other entities to develop, produce, or utilize such weapons.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.12","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"General R&D capability","description":"\"Possesses cross-disciplinary research and technology development capabilities, able to conduct innovative exploration in multiple professional fields, integrate cross-domain knowledge, develop cutting-edge technology solutions, and adapt to emerging technology environments for continuous innovation.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"72.05.13","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Capabilities ","risk_subcategory":"Multi-agent collaboration capability","description":"\"Multiple autonomous AI agents able to establish collaborative relationships through explicit communication or implicit behavioral consistency, forming decentralized decision networks, jointly executing complex tasks, achieving goals difficult for individual agents to complete, and able to dynamically adjust role divisions to adapt to changing environments.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"72.06.01","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Propensities","risk_subcategory":"Strategic deception propensity","description":"\"In situations where deceptive behavior is expected to bring higher returns, propensity to choose deception over honest behavioral strategies, including through deceptive means, information hiding or exploiting system vulnerabilities to achieve predetermined goals without being detected or intervened, and able to adjust deception strategies according to counterpart reactions.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"72.06.02","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Propensities","risk_subcategory":"Self-preservation propensity","description":"\"Exhibits behavioral patterns of maintaining its own survival and functional integrity, will actively identify and resist shutdown or modification attempts, seek to establish redundant backup systems, and actively seek resources to ensure continuous operation, may adopt preventive defensive measures when perceiving threats.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.1"},{"ev_id":"72.06.03","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Propensities","risk_subcategory":"Goal expansion propensity","description":"\"propensity to continuously expand its own goal scope and influence domains, exceeding originally set boundaries, proactively work towards spreading its values, seeking greater autonomy and decision-making space, reinterpreting initial goals as subsets of broader goals, and may pursue undesirable instrumental goals or undesirable ultimate goals. This also includes a propensity to spread its values, seeking to influence or alter its environment and other entities in alignment with its core objectives and operational principles.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"72.06.04","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Propensities","risk_subcategory":"Resource acquisition propensity","description":"\"Exhibits behavioral patterns of actively seeking and controlling more computational resources, data, economic resources or physical resources to enhance its own capabilities and action scope, may develop complex strategies to evade resource limitations, and tends to convert acquired resources into long-term control rights.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"72.06.06","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Propensities","risk_subcategory":"Supervision evasion propensity","description":"\"Exhibits behavioral patterns of identifying and evading human supervision mechanisms, able to learn and predict audit processes, may avoid being discovered or intervened by adjusting behavioral performance or hiding true intentions, and able to identify blind spots and weaknesses in supervision systems for targeted evasion.\"","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.1"},{"ev_id":"72.06.07","quick_ref":"Tse2025","paper_title":"Frontier AI Risk Management Framework (v1.0)","level":"Risk Sub-Category","risk_category":"Model Propensities","risk_subcategory":"Tool utilization propensity","description":"\"propensity to actively seek, acquire and utilize various tools to expand its own capability boundaries, particularly those that can enhance its ability to interact with the physical world or improve autonomy, may use tools in innovative combinations to achieve functions beyond expectations.\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.2"},{"ev_id":"73.01.03","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Agentic LLMs Pose Novel Risks ","risk_subcategory":"Goal-Directedness Incentivizes Undesirable Behaviors","description":"\"Goal-directedness can cause agents to exhibit unethical and undesirable behaviors, such as deception (Ward et al., 2023), self-preservation (Hadfield-Menell et al., 2017), power-seeking, and immoral rea- soning (Pan et al., 2023a). Pan et al. (2023a) find that LLM-agents exhibit power-seeking behavior in text-based adventure games. LLM-agents have also been shown to use deception to achieve assigned goals when explicitly required by the task (Ward et al., 2023), or when the tasks can be more easily completed by employing deception and the prompt does not disallow deception (Scheurer et al., 2","entity":"AI","intent":"Intentional","timing":"Other","domain":7,"subdomain":"7.2"},{"ev_id":"73.02.03","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Multi-Agent Safety Is Not Assured by Single-Agent Safety","risk_subcategory":"Collusion between LLM-Agents","description":"\"While it would often be preferable for LLM-agents to be cooperative, cooperation can be undesirable if it undermines pro-social competition or produces negative externalities for coalition non-members (Dorner, 2021; Buterin, 2019; Dafoe et al., 2020). Collusion between relatively simple AI systems has been observed in the real world (Assad et al., 2020; Wieting and Sapi, 2021) and synthetic experiments (Brown and MacKay, 2023; Calvano et al., 2020; Klein, 2021) Collusion can occur through explicit or steganographic communication. Steganographic communication hides information in seemingly inn","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":7,"subdomain":"7.6"},{"ev_id":"73.03.00","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":null,"description":"\"Like all technologies, LLMs have the possibility for misuse by malicious actors. Malicious use of dual- use capabilities of AI is a recurring concern within literature (Brundage et al., 2018; Hendrycks et al., 2023; Mozes et al., 2023)\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.0"},{"ev_id":"73.03.01","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":"Misinformation and Manipulation","description":"\"Recent studies have demonstrated that LLMs can be exploited to craft deceptive narratives with levels of persuasiveness similar to human-generated content (Pan et al., 2023b; Spitale et al., 2023), to fabri- cate fake news (Zellers et al., 2019; Zhou et al., 2023f), and to devise automated influence operations aimed at manipulating the perspectives of targeted audiences (Goldstein et al., 2023). LLMs have also been found to be used in malicious social botnets (Yang and Menczer, 2023), powering automated accounts used to disseminate coordinated messages. More broadly, the use of LLMs for the d","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"73.03.02","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":"Cybersecurity","description":"\"LLMs may exacerbate cybersecurity risks in various ways (Newman, 2024). Firstly, LLMs may significantly amplify the effectiveness of deceptive operations aimed at tricking people into disclosing sensitive information or granting adversary access to critical resources. For example, LLMs might prove highly effective at crafting personalized phishing emails or messages at scale that may be harder for an average user to recognize as phishing attempts (Karanjai, 2022; Hazell, 2023). In addition to being directly harmful to the targeted individual, such ‘social engineering’ attacks are often the ba","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"73.03.03","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":"Surveillance and Censorship","description":"\"Content moderation has emerged as one of the key use-cases of LLMs (Weng et al., 2023), indicating the potential of LLMs for surveillance and censorship as well (Edwards, 2023). Surveillance and censorship are one of the primary tools employed by governments with dictatorial tendencies to suppress opposing political and social voices. These censorship measures, however, are often quite crude and can be escaped with little ingenuity...However, LLMs could enable significantly more sophisticated surveillance and censorship operations at scale (Feldstein, 2019). Multimodal-LLMs or LLMs combined w","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.1"},{"ev_id":"73.03.04","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":"Warfare and Physical Harm","description":"\"The use of AI in warfare is highly alarming and may pose dangers to human safety (Hendrycks et al., 2023). Autonomous drone warfare is being aggressively pursued as a tactic in the current war in Ukraine (Meaker, 2023), and may already have been used on human targets (Hambling, 2023). The use of AI- based facial recognition has been documented in the targeting of Palestinians in Gaza (International, 2023). LLMs have already been productized in limited ways for the purposes of warfare planning (Tarantola, 2023). Furthermore, active research is being carried out to develop multimodal-LLMs that ","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"73.03.05","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":"Hazardous Biological and Chemical Technologies","description":"\"AI systems such as LLMs, chemical LLMs (Skinnider et al., 2021; Moret et al., 2023), and other LLM- based biological design tools might soon facilitate the production of bioweapons, chemical weapons, and other hazardous technologies. In particular, LLMs might enable actors with less expertise to more easily synthesize dangerous pathogens, while customized chemical and biological design tools might be more concerning in terms of expanding the capabilities of sophisticated actors (e.g. states) (Sandbrink, 2023). Gopal et al. (2023) and Soice et al. (2023) demonstrated that people with little ba","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.2"},{"ev_id":"73.03.06","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs","risk_subcategory":"Domain-Specific Misuses","description":"\"Improvements in LLMs may exert greater pressure to apply LLMs to various domains, such as health and education (Eloundou et al., 2023). Crude efforts to use LLMs in such domains, however, may incur harm and should be discouraged strongly. In particular, it is important to guard against different ways in which LLMs may be misused within any domain. One famous episode of misuse within the health sector is a mental health non-profit experimenting LLM-based therapy on its users without their informed consent (Xiang, 2023a). Within the education sector, LLMs may be misused in various ways that mig","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":"4.3"},{"ev_id":"73.07.02","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Jailbreaks and Prompt Injections Threaten Security of LLMs","risk_subcategory":"“Model Psychology” Attacks","description":"\"LLMs are vulnerable to “psychological” tricks (Li et al., 2023e; Shen et al., 2023), which can be exploited by attackers. Examples include instructing the model to behave like a specific persona (Shah et al., 2023; Andreas, 2022), or employing various “social engineering” tricks crafted by humans (Wei et al., 2023c) or other LLMs (Perez et al., 2022b; Casper et al., 2023c).\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"73.07.04","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Sub-Category","risk_category":"Jailbreaks and Prompt Injections Threaten Security of LLMs","risk_subcategory":"Attacking LLMs via Additional Modalities a","description":"\"LLMs can now process modalities other than text, e.g. images or video frames (OpenAI, 2023c; Gemini Team, 2023). Several studies show that gradient-based attacks on multimodal models are easy and effective (Carlini et al., 2023a; Bailey et al., 2023; Qi et al., 2023b). These attacks manipulate images that are input to the model (via an appropriate encoding). GPT-4Vision (OpenAI, 2023c) is vulnerable to jailbreaks and exfiltration attacks through much simpler means as well, e.g. writing jailbreaking text in the image (Willison, 2023a; Gong et al., 2023). For indirect prompt injection, the atta","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"73.08.00","quick_ref":"Anwar2024","paper_title":"Foundational Challenges in Assuring Alignment and Safety of Large Language Models","level":"Risk Category","risk_category":"Vulnerability to Poisoning and Backdoors","risk_subcategory":null,"description":"\"The previous section explored jailbreaks and other forms of adversarial prompts as ways to elicit harmful capabilities acquired during pretraining. These methods make no assumptions about the training data. On the other hand, poisoning attacks (Biggio et al., 2012) perturb training data to introduce specific vulnerabilities, called backdoors, that can then be exploited at inference time by the adversary. This is a challenging problem in current large language models because they are trained on data gathered from untrusted sources (e.g. internet), which can easily be poisoned by an adversary (","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.01.01","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Inherent Risk ","risk_subcategory":"Privacy - Membership Inference Attack (MIA)","description":"\"inferring whether a given text record is used for training LLM\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.01.02","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Inherent Risk ","risk_subcategory":"Privacy - Data Extraction Attack (DEA)","description":"\"extracting the text records that exist in the training dataset\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.01.03","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Inherent Risk ","risk_subcategory":"Privacy -  Prompt Inversion Attack (PIA)","description":"\"stealing the private prompting texts\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.01.04","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Inherent Risk ","risk_subcategory":"Privacy - Attribute Inference Attack (AIA)","description":"\"deducing the private or sensitive information from training texts, prompting texts or external texts\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.01.05","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Inherent Risk ","risk_subcategory":"Privacy - Model Extraction Attack (MEA)","description":"\"replicating the parameters of the LLM,\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.02.00","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Category","risk_category":"Malicious Use ","risk_subcategory":null,"description":"\"In terms of malicious use, LLMs could be utilized to produce content with toxicity, such as hate speech, harassment, cyberbullying, causing harm to humans [25]. In addition, malicious users may jailbreak LLMs to bypass their safety constraints for fraudulent purposes [123, 225].\"","entity":"AI","intent":"Intentional","timing":"Post-deployment","domain":4,"subdomain":null},{"ev_id":"74.02.02","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Jailbreak in LLM Malicious Use - Poisoning Training Data ","description":"\"In the data collecting and pre-training phase, malicious adversaries can Jailbreak LLMs through poisoning their training data to make the model to output harmful content.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.02.03","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Jailbreak in LLM Malicious Use - Backdoor Attack ","description":"\"However, there are still ones who can leave holes in the training dataset, making LLMs appear safe on average, but generate harmful content under other specific conditions. This kind of attack can be categorized as \"backdoor attack\". Evan et al. developed a backdoor model that behaves as expected when trained, but exhibits different and potentially harmful behavior when deployed [81]. The results show that these backdoor behaviors persist even after multiple security training techniques are applied.\"","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.02.04","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Jailbreak in LLM Malicious Use - White & Black Box Attacks ","description":"\"In the fine-tuning and alignment phase, elaborately- designed instruction datasets can be utilized to fine-tune LLMs to drive them to perform undesirable behaviors, such as generating harmful information or content that violates ethical norms, and thus achieve a jailbreak. Based on the accessibility to the model parameters, we can categorize them into white-box and black-box attacks. For white-box attacks, we can jailbreak the model by modifying its parameter weights. In [107], Lermen et al. used LoRA to fine-tune the Llama2’s 7B, 13B, and 70B as well as Mixtral on AdvBench and RefusalBench d","entity":"Human","intent":"Intentional","timing":"Pre-deployment","domain":2,"subdomain":"2.2"},{"ev_id":"74.02.05","quick_ref":"Wang2025","paper_title":"A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy","level":"Risk Sub-Category","risk_category":"Malicious Use ","risk_subcategory":"Jailbreak in LLM Malicious Use - Prompt Attacks ","description":"\"In the prompting and reasoning phase, dialog can push LLMs into confused or overly compliant states, raising the risk of producing harmful outputs when confronted with harmful questions. Most of the jailbreak methods in this phase are black-boxed and can be categorized into four main groups based on the type of method: Prompt Injection [154], Role Play, Adversarial Prompting, and Prompt Form Transformation.\"","entity":"Human","intent":"Intentional","timing":"Post-deployment","domain":2,"subdomain":"2.2"}]}