MIT AI Risk Repository

Browse AI risks

554 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

554 entries · page 10 of 12

  1. From Future Risks of Frontier AI (GOS2023)

  2. From Future Risks of Frontier AI (GOS2023)

  3. "The operational design domain (ODD) is a technical description of the application’s operational environment, initially conceptualized for autonomous driving systems. An inadequate specification of the ODD limits essential functions such as testing the learned functionality and out-of-distribution detection."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  4. "The expected performance of the AI system should be planned adequately. Hereby, an important aspect is that chosen performance metrics are meaningful for presenting the intended functionality. Otherwise, expectations and safety requirements can be unfulfillable at later life cycle stages."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  5. 59.11.00 · Risk Category

    Incorrect data labels

    "Data labels are essential for any supervised learning algorithm since they preset the result of the learning process. If the correctness of the data labels is not given, the AI system is prevented from learning the ground truth and therefore the intended functionality."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  6. "The model specifications have significant impact on the functionality of an AI system. The developer mak- ing wrong decisions might cause the AI system to behave biased and unreliable."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  7. "AI systems tend to show unreliable behavior when confronted with rare or ambiguous input data, also called corner cases. Therefore, the controlled behavior is required whenever the AI system is faces a corner case."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  8. 59.20.00 · Risk Category

    Lack of robustness

    "Robustness characterizes the resilience of an AI system’s output against minor changes in the input domain. A great variation in an AI system’s response to small input changes indicates unreliable outputs."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  9. "Until the deployment of the AI application into its operational environment, the AI system has been tested with a test set that aims to approximate the distribution of operational data. However, an unexpected deviation in this approximation can cause an AI application to behave unreliably. Therefore, its behavior under confrontation with operational data needs to be evaluated."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  10. 59.26.01 · Risk Sub-Category

    Mode

    Technical

    "Technical AI hazards are the root causes of technical deficiencies in the AI system. An example of such an AI hazard is overfitting, which describes a model’s excessive adaptation to the training dataset. Quantitative methods to assess (metrics) and treat (mitigation means) exist for technical AI hazards, which might be performed automatically. In case of overfitting, metrics are based on the comparison of performance between the training and validation datasets, and mitigation means may include regularization techniques, among others."

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  11. 59.26.03 · Risk Sub-Category

    Mode

    Procedural

    "The third class encompasses procedural AI hazards. These pertain to issues arising from processes and actions made by individuals involved in the develop- ment process. Such hazards are not readily quantifiable and necessitate alter- native mitigation strategies. An example of such an AI hazard would be ”poor model design choices,” which could be expressed, for instance, through a devel- oper’s decision to select an unsuitable AI model for a given problem. Due to the challenges in quantifying and mitigating these issues, qualitative approaches must be employed. In the case of the aforemention

    From AI Hazard Management: A Framework for the Systematic Management of Root Causes for AI Risks (Schnitzer2024)

  12. 60.02.01 · Risk Sub-Category

    Risks from malfunctions

    Reliability issues

    "Relying on general-purpose AI products that fail to fulfil their intended function can lead to harm. For example, general- purpose AI systems can make up facts (‘hallucination’), generate erroneous computer code, or provide inaccurate medical information. This can lead to physical and psychological harms to consumers and reputational, financial and legal harms to individuals and organisations."

    From International AI Safety Report 2025 (Bengio2025)

  13. 61.02.31 · Risk Sub-Category

    Sources of systemic risks from general-purpose AI

    Lack of ability to generate accurate information

    "AI models may generate false or misleading information due to their lack of capability in discerning truth."

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

  14. 61.02.32 · Risk Sub-Category

    Sources of systemic risks from general-purpose AI

    Lack of ethical decision-making

    "AI models and systems that lack moral reasoning capabilities may make decisions that are unethical or harmful."

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

  15. 61.02.46 · Risk Sub-Category

    Sources of systemic risks from general-purpose AI

    Unclear attribution from AI component interactions

    "Interactions between different AI components can cause harm, but it may be difficult to pinpoint which components are the cause."

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

  16. 62.14.02 · Risk Sub-Category

    Model Development

    Data-related (Lack of cross-organizational documentation)

    "When sharing data between multiple organizations, documentation may be missing or inadequate, making it difficult for other organizations to understand it. For example, a lack of metadata or a change in schema by a collaborating party can result in an unusable dataset and wasted data collection efforts, or it can lead to misunderstandings about the dataset’s limitations, resulting in downstream risks related to its use [173]."

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

  17. 62.14.03 · Risk Sub-Category

    Model Development

    Data-related (Manipulation of data by non-domain experts)

    "Manipulating data (e.g., training data) carries a set of assumptions on how the data should appear and be used by those performing the manipulation. Common manipulations applied on data in the context of AI models include defining the ground truth label and merging different data formats or sources. People who have little or no expertise in the domain of the data performing such manipulations may render the data unusable or harmful to the development of the AI system [173]."

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

  18. 62.15.00 · Risk Sub-Category

    Model Development

    Training-related (Robust overfitting in adversarial training)

    "Adversarial training can be affected by robust overfitting, where the model’s robustness on test data decreases during further training, particularly after the learning rate decay. This issue has been consistently observed across various datasets and algorithms in adversarial training settings [163, 230]. Robust over- fitting can affect the model’s ability to generalize effectively and reduce its resilience to adversarial attacks."

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

  19. 62.15.02 · Risk Sub-Category

    Model Development

    Training-related (Poor model confidence calibration)

    "Models can be affected by poor confidence calibration [85], where the predicted probabilities do not accurately reflect the true likelihood of ground truth cor- rectness. This miscalibration makes it difficult to interpret the model’s predic- tions reliably, as high accuracy does not guarantee that the confidence levels are meaningful. This can cause overconfidence in incorrect predictions or un- derconfidence in correct ones."

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

  20. 62.15.10 · Risk Sub-Category

    Model Development

    Fine-tuning related (Catastrophic forgetting due to continual instruction fine-tuning)

    "Catastrophic forgetting occurs when a model loses its ability to retain previously learned tasks (or factual information) after being trained on new ones. In language models, this can occur due to continual instruction tuning. This tendency may become more pronounced as the model’s size increases [127]."

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

  21. 62.16.07 · Risk Sub-Category

    Model Evaluations

    General Evaluations (AI outputs for which evaluation is too difficult for humans)

    "When AI models are trained through evaluation with human feedback, such as reinforcement learning from human feedback, their outputs can be challenging to assess, as they may contain hard-to-detect errors or issues that only become apparent over time. The human evaluator can rate incorrect outputs positively or similar to correct outputs. This can lead to the model learning to produce subtly incorrect or harmful outputs, such as code with software vulnerabilities, or politically biased information. In extreme cases where a model is deceiving users, complicated outputs can contain hidden error

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

  22. 62.19.08 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Models distracted by irrelevant context

    "Models can easily become distracted by irrelevant provided information (such as “context” in LLMs), leading to a significant decrease in their performance after introducing irrelevant information. This can happen with different prompting techniques, including chain-of-thought prompting [184]."

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

  23. 62.19.09 · Risk Sub-Category

    Attacks on GPAIs/GPAI Failure Modes

    Knowledge conflicts in retrieval-augmented LLMs

    "AI models can be particularly sensitive to coherent external evidence, even when they come into conflict with the models’ prior knowledge. This may lead to models producing false outputs given false information during the retrieval- augmentation process, despite only a relatively small amount of false informa- tion input that is inconsistent with the model’s prior knowledge trained on much larger amounts of data [220]."

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

  24. 62.30.01 · Risk Sub-Category

    Impacts of AI (Physical)

    Damage to critical infrastructure

    "The integration of AI systems within critical infrastructure, ranging from trans- portation to power systems, can cause substantial damage in cases of failure or malfunction. With the increasing number of Internet of Things (IoT) devices and interconnected cyber-physical systems, critical infrastructure becomes even more vulnerable [171, 174]."

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

  25. 62.30.04 · Risk Sub-Category

    Impacts of AI (Physical)

    AI Systems interacting with brittle environments

    "Deployed AI systems can rely on physical sensors and data sources that may exhibit hardware drift and thus data distribution drift over time. This distribu- tion drift may affect system robustness and performance. This usually involves AI systems working in undigitized and physical environments."

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

  26. 62.32.04 · Risk Sub-Category

    Impacts of AI (Cyberattacks)

    Models generating code with security vulnerabilities

    "Models can generate code or coding suggestions that contain security vulner- abilities. This may occur across various LLM-based model families, including more advanced models with superior coding performance, where the tendency to produce insecure code is even more pronounced [26]."

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

  27. 65.02.01 · Risk Sub-Category

    Training Data Risks (Data laws)

    Data usage restrictions

    "Laws and other restrictions can limit or prohibit the use of some data for specific AI use cases."

    From AI Risk Atlas (IBM2025)

  28. 65.02.02 · Risk Sub-Category

    Training Data Risks (Data laws)

    Data acquisition restrictions

    "Laws and other regulations might limit the collection of certain types of data for specific AI use cases."

    From AI Risk Atlas (IBM2025)

  29. 65.02.03 · Risk Sub-Category

    Training Data Risks (Data laws)

    Data transfer restrictions

    "Laws and other restrictions can limit or prohibit transferring data."

    From AI Risk Atlas (IBM2025)

  30. 65.06.01 · Risk Sub-Category

    Training Data Risks (Accuracy)

    Data contamination

    "Data contamination occurs when incorrect data is used for training. For example, data that is not aligned with model’s purpose or data that is already set aside for other development tasks such as testing and evaluation."

    From AI Risk Atlas (IBM2025)

  31. 65.06.02 · Risk Sub-Category

    Training Data Risks (Accuracy)

    Unrepresentative data

    "Unrepresentative data occurs when the training or fine-tuning data is not sufficiently representative of the underlying population or does not measure the phenomenon of interest."

    From AI Risk Atlas (IBM2025)

  32. 65.07.01 · Risk Sub-Category

    Training Data Risks (Value alignment)

    Improper retraining

    "Using undesirable output (for example, inaccurate, inappropriate, and user content) for retraining purposes can result in unexpected model behavior."

    From AI Risk Atlas (IBM2025)

  33. 65.07.02 · Risk Sub-Category

    Training Data Risks (Value alignment)

    Improper data curation

    "Improper collection and preparation of training or tuning data includes data label errors and by using data with conflicting information or misinformation."

    From AI Risk Atlas (IBM2025)

  34. 65.13.01 · Risk Sub-Category

    Inference risks (Accuracy)

    Poor model accuracy

    "Poor model accuracy occurs when a model’s performance is insufficient to the task it was designed for. Low accuracy might occur if the model is not correctly engineered, or there are changes to the model’s expected inputs."

    From AI Risk Atlas (IBM2025)

  35. 65.15.01 · Risk Sub-Category

    Output risks (Value alignment)

    Incomplete advice

    "When a model provides advice without having enough information, resulting in possible harm if the advice is followed."

    From AI Risk Atlas (IBM2025)

  36. 66.04.04 · Risk Sub-Category

    Societal and Cultural

    Productivity loss

    "End user's loss of productivity due to the underperfomance of a genAI application, including producing nonsensical or poor quality outputs, degrading its utility."

    From A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents (Li2025)

  37. "The chatbot makes a deal, commitment, or other consequential action with its output that the deployer did not intend."

    From Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024)

  38. "The chatbot gives guidance that ranges from simply unhelpful to harmful if acted on."

    From Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024)

  39. 69.04.02 · Risk Sub-Category

    Bad advice/failure to generate helpful content

    Unhelpful responses

  40. 69.04.03 · Risk Sub-Category

    Bad advice/failure to generate helpful content

    Bad links and references

  41. 69.04.04 · Risk Sub-Category

    Bad advice/failure to generate helpful content

    Nonsensical content

  42. 70.01.02 · Risk Sub-Category

    Physical Risks

    Accidental harm

    "Automation in sectors ranging from manufacturing to healthcare has and will increasingly put humans into close contact with EAI systems [7]. This interaction increases the risk of accidental physical harm. Though accidental harm has been a longstanding issue in industrial robotics, increased AI capabilities could exacerbate this risk; several recent reports document an increase in industrial injuries following the introduction of AI-controlled robots [66–68]."

    From Embodied AI: Emerging Risks and Opportunities for Policy Action (Perlo2025)

  43. 71.01.04 · Risk Sub-Category

    Scientific Domain of Agents

    Physical (Mechanical ) Risks

    "Physical (mechanical) risks are associated with robotics and automated systems, which could lead to equipment malfunctions or physical harm in laboratory settings."

    From Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy (Tang2025)

  44. 72.03.00 · Risk Category

    Accident Risks

    "Risks arising from operational failures, model misjudgments, or improper human operation of AI systems deployed in safety-critical infrastructure, where single points of failure can trigger cascading catastrophic consequences."

    From Frontier AI Risk Management Framework (v1.0) (Tse2025)

  45. 72.03.01 · Risk Sub-Category

    Accident Risks

    Nuclear Power Systems

    "General-purpose AI deployed for reactor monitoring, control system optimization, or emergency response coordination could misinterpret sensor data, fail to recognize critical safety conditions, or make erroneous control decisions during emergency scenarios. Given the catastrophic potential of nuclear accidents, even minor AI reasoning errors in safety-critical functions could lead to core meltdowns, radiation releases, or widespread contamination affecting hundreds of thousands of people across international borders."

    From Frontier AI Risk Management Framework (v1.0) (Tse2025)

  46. 72.03.03 · Risk Sub-Category

    Accident Risks

    Other Critical Infrastructure Control Systems

    "General-purpose AI deployed in power grid management, water treatment facilities, telecommunications networks, or transportation coordination systems could misinterpret operational data, fail to anticipate cascading failure modes, or make control decisions that destabilize interconnected infrastructure networks. Infrastructure failures could result in widespread blackouts, contaminated water supplies, communications breakdowns, and the collapse of essential services supporting hundreds of thousands of people."

    From Frontier AI Risk Management Framework (v1.0) (Tse2025)

  47. 06.06.00 · Risk Category

    Lack of transparency

    "The idea of a "black box" making decisions without any explanation, without offering insight in the process, has a couple of disadvantages: it may fail to gain the trust of its users and it may fail to meet regulatory standards such as the ability to audit."

    From A framework for ethical Ai at the United Nations (Hogenhout2021)

  48. 10.05.00 · Risk Category

    Lack of transparency

    "In situations in which the development and use of AI are not explained to the user, or in which the decision processes do not provide the criteria or steps that constitute the decision, the use of AI becomes inexplicable."

    From Social Impacts of Artificial Intelligence and Mitigation Recommendations: An Exploratory Study (Paes2023)

  49. "Transparency is the characteristic of a system that describes the degree to which appropriate information about the system is communicated to relevant stakeholders, whereas explainability describes the property of an AI system to express important factors influencing the results of the AI system in a way that is understandable for humans....Information about the model underlying the decision-making process is relevant for transparency. Systems with a low degree of transparency can pose risks in terms of their fairness, security and accountability. "

    From Sources of Risk of AI Systems (Steimers2022)

  50. The ability to explain the outputs to users and reason correctly

    From Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment (Liu2024)

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