MIT AI Risk Repository

Browse AI risks

977 risk entries extracted from 74 frameworks, coded by domain, subdomain, causal entity, intent and timing. Filter, then export the current selection with its licence and citation attached.

Reset

977 entries · page 5 of 20

  1. 58.06.11 · Risk Sub-Category

    Human rights and civil liberties

    Privacy loss

    "Privacy loss - Unwarranted exposure of an individual’s private life or personal data through cyberattacks, doxxing, etc."

    From A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms (Abercrombie2024)

  2. 61.02.11 · Risk Sub-Category

    Sources of systemic risks from general-purpose AI

    Centralized platforms deployed at scale

    "The widespread use of common AI platforms can create centralized points of failure, making systems more vulnerable to disruptions or attacks"

    From A Taxonomy of Systemic Risks from General-Purpose AI (Uuk2025)

  3. 61.02.33 · Risk Sub-Category

    Sources of systemic risks from general-purpose AI

    Limitations in adversarial robustness

    "AI models and systems are vulnerable to manipulation through adversarial inputs."

    From A Taxonomy of Systemic Risks from General-Purpose AI (Uuk2025)

  4. 62.14.05 · Risk Sub-Category

    Model Development

    Training-related (Adversarial examples)

    "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."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  5. 62.19.01 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Jailbreak of a model to subvert intended behavior

    "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]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  6. 62.19.02 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Jailbreak of a multimodal model

    "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]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  7. 62.19.03 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Transferable adversarial attacks from open to closed-source mod- els

    "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]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  8. 62.19.05 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Text encoding-based attacks

    "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]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  9. 62.19.07 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Vulnerabilities to jailbreaks exploiting long context windows (many- shot jailbreaking)

    "Language models with long context windows are vulnerable to new types of ex- ploitations that are ineffective on models with shorter context windows. While few-shot jailbreaking, which involves providing few examples of the desired harmful output, might not trigger a harmful response, many-shot jailbreak- ing, which involves a higher number of such examples, increases the likelihood of eliciting an undesirable output. These vulnerabilities become more significant as context windows expand with newer model releases [7]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  10. 62.19.12 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Misuse of AI model by user-performed persuasion

    "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]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  11. 62.27.01 · Risk Sub-Category

    Deployment (Model Release)

    Non-decomissionability of models with open weights

    "If the model parameter weights are released or leaked in a security breach, the model cannot be decommissioned because the developer no longer has control over the publicly available model or its use. This prevents effective management and control of an open-sourced or leaked model. Models with publicly available weights are also easier to reconfigure, enabling misuse [178]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  12. 62.28.01 · Risk Sub-Category

    Cybersecurity

    Interconnectivity with malicious external tools

    "The growing integration and interconnectivity with external tools and plugins increase the risk of exposure to malicious external inputs. This interconnectivity makes it easier for external tools to introduce harmful content [220]."

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  13. 62.28.04 · Risk Sub-Category

    Cybersecurity

    Model weight leak

    "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

    From Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024)

  14. "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."

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  15. "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"

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  16. "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

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  17. "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)."

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  18. "Steganography is the practice of hiding coded messages in GenAI model outputs, which may allow malicious actors to communicate covertly.8"

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  19. "Privacy Compromise attacks reveal sensitive or private information that was used to train a model. For example, personally identifiable information or medical records."

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  20. "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)."

    From Generative AI Misuse: A Taxonomy of Tactics and Insights from Real-World Data (Marchal2024)

  21. 65.09.01 · Risk Sub-Category

    Inference risks (Robustness)

    Prompt injection attack

    "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."

    From AI Risk Atlas (IBM2025)

  22. 65.09.02 · Risk Sub-Category

    Inference risks (Robustness)

    Extraction attack

    "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."

    From AI Risk Atlas (IBM2025)

  23. 65.09.03 · Risk Sub-Category

    Inference risks (Robustness)

    Evasion attack

    "Evasion attacks attempt to make a model output incorrect results by slightly perturbing the input data that is sent to the trained model."

    From AI Risk Atlas (IBM2025)

  24. 65.10.01 · Risk Sub-Category

    Inference risks (Multi-category)

    Jailbreaking

    "A jailbreaking attack attempts to break through the guardrails that are established in the model to perform restricted actions."

    From AI Risk Atlas (IBM2025)

  25. 65.10.02 · Risk Sub-Category

    Inference risks (Multi-category)

    Prompt priming

    "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."

    From AI Risk Atlas (IBM2025)

  26. 65.11.01 · Risk Sub-Category

    Inference risks (Privacy)

    Membership inference attack

    "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."

    From AI Risk Atlas (IBM2025)

  27. 65.11.02 · Risk Sub-Category

    Inference risks (Privacy)

    Attribute inference attack

    "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."

    From AI Risk Atlas (IBM2025)

  28. 65.15.02 · Risk Sub-Category

    Output risks (Value alignment)

    Harmful code generation

    "Models might generate code that causes harm or unintentionally affects other systems."

    From AI Risk Atlas (IBM2025)

  29. 73.07.02 · Risk Sub-Category

    Jailbreaks and Prompt Injections Threaten Security of LLMs

    “Model Psychology” Attacks

    "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)."

    From Foundational Challenges in Assuring Alignment and Safety of Large Language Models (Anwar2024)

  30. 73.07.04 · Risk Sub-Category

    Jailbreaks and Prompt Injections Threaten Security of LLMs

    Attacking LLMs via Additional Modalities a

    "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

    From Foundational Challenges in Assuring Alignment and Safety of Large Language Models (Anwar2024)

  31. 74.01.01 · Risk Sub-Category

    Inherent Risk

    Privacy - Membership Inference Attack (MIA)

    "inferring whether a given text record is used for training LLM"

    From A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025)

  32. 74.01.02 · Risk Sub-Category

    Inherent Risk

    Privacy - Data Extraction Attack (DEA)

    "extracting the text records that exist in the training dataset"

    From A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025)

  33. 74.01.03 · Risk Sub-Category

    Inherent Risk

    Privacy - Prompt Inversion Attack (PIA)

    "stealing the private prompting texts"

    From A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025)

  34. 74.01.04 · Risk Sub-Category

    Inherent Risk

    Privacy - Attribute Inference Attack (AIA)

    "deducing the private or sensitive information from training texts, prompting texts or external texts"

    From A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025)

  35. 74.01.05 · Risk Sub-Category

    Inherent Risk

    Privacy - Model Extraction Attack (MEA)

    "replicating the parameters of the LLM,"

    From A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025)

  36. 74.02.05 · Risk Sub-Category

    Malicious Use

    Jailbreak in LLM Malicious Use - Prompt Attacks

    "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."

    From A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025)

  37. "These risks arise from the LM outputting false, misleading, nonsensical or poor quality information, without malicious intent of the user. (The deliberate generation of "disinformation", false information that is intended to mislead, is discussed in the section on Malicious Uses.) Resulting harms range from unintentionally misinforming or deceiving a person, to causing material harm, and amplifying the erosion of societal distrust in shared information. Several risks listed here are well-documented in current large-scale LMs as well as in other language technologies"

    From Taxonomy of Risks posed by Language Models (Weidinger2022)

  38. "Harms that arise from the language model providing false or misleading information"

    From Ethical and social risks of harm from language models (Weidinger2021)

  39. "AI systems generating and facilitating the spread of inaccurate or misleading information that causes people to develop false beliefs"

    From Sociotechnical Safety Evaluation of Generative AI Systems (Weidinger2023)

  40. 02.02.00 · Risk Category

    Untruthful Content

    "The LLM-generated content could contain inaccurate information"

    From Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  41. 02.02.01 · Risk Sub-Category

    Untruthful Content

    Factuality Errors

    "The LLM-generated content could contain inaccurate information" which is factually incorrect

    From Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  42. 02.09.00 · Risk Category

    Hallucinations

    "LLMs generate nonsensical, untruthful, and factual incorrect content"

    From Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  43. 02.09.05 · Risk Sub-Category

    Hallucinations

    Pursuing Consistent Context

    "LLMs have been demonstrated to pursue consistent context [129]–[132], which may lead to erroneous generation when the prefixes contain false information. Typical examples include sycophancy [129], [130], false demonstrations-induced hallucinations [113], [133], and snowballing [131]. As LLMs are generally fine-tuned with instruction-following data and user feedback, they tend to reiterate user-provided opinions [129], [130], even though the opinions contain misinformation. Such a sycophantic behavior amplifies the likelihood of generating hallucinations, since the model may prioritize user op

    From Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)

  44. 03.02.00 · Risk Category

    Hallucinations

    "The inclusion of erroneous information in the outputs from AI systems is not new. Some have cautioned against the introduction of false structures in X-ray or MRI images, and others have warned about made-up academic references. However, as ChatGPT-type tools become available to the general population, the scale of the problem may increase dramatically. Furthermore, it is compounded by the fact that these conversational AIs present true and false information with the same apparent “confidence” instead of declining to answer when they cannot ensure correctness. With less knowledgeable people,

    From Navigating the Landscape of AI Ethics and Responsibility (Cunha2023)

  45. 04.05.00 · Risk Category

    Misleading Information

    Large models are usually susceptible to hallucination problems, sometimes yielding nonsensical or unfaithful data that results in misleading outputs.

    From Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements (Deng2023)

  46. 05.04.00 · Risk Category

    Hallucinations

    Significant concerns are raised about LLMs inadvertently generating false or misleading information, as well as erroneous code. Papers not only critically analyze various types of reasoning errors in LLMs but also examine risks associated with specific types of misinformation, such as medical hallucinations. Given the propensity of LLMs to produce flawed outputs accompanied by overconfident rationales and fabricated references, many sources stress the necessity of manually validating and fact-checking the outputs of these models.

    From Mapping the Ethics of Generative AI: A Comprehensive Scoping Review (Hagendorff2024)

  47. 11.05.01 · Risk Sub-Category

    Societal System Harms

    Information harms

    information-based harms capture concerns of misinformation, disinformation, and malinformation. Algorithmic systems, especially generative models and recommender, systems can lead to these information harms

    From Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction (Shelby2023)

  48. 16.03.01 · Risk Sub-Category

    Risk area 3: Misinformation Harms

    Disseminating false or misleading information

    "Where a LM prediction causes a false belief in a user, this may threaten personal autonomy and even pose downstream AI safety risks [99]."

    From Taxonomy of Risks posed by Language Models (Weidinger2022)

  49. 16.03.02 · Risk Sub-Category

    Risk area 3: Misinformation Harms

    Causing material harm by disseminating false or poor information e.g. in medicine or law

    "Induced or reinforced false beliefs may be particularly grave when misinformation is given in sensitive domains such as medicine or law. For example, misin- formation on medical dosages may lead a user to cause harm to themselves [21, 130]. False legal advice, e.g. on permitted owner- ship of drugs or weapons, may lead a user to unwillingly commit a crime. Harm can also result from misinformation in seemingly non-sensitive domains, such as weather forecasting. Where a LM prediction endorses unethical views or behaviours, it may motivate the user to perform harmful actions that they may otherw

    From Taxonomy of Risks posed by Language Models (Weidinger2022)

Informational only, not legal advice. Verify every claim against the linked official sources and consult qualified counsel before acting.