MIT AI Risk Repository

Browse AI risks

157 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 Also filtered by framework IBM2025 ×

157 entries · page 1 of 4

  1. 65.04.01 · Risk Sub-Category

    Training Data Risks (Fairness)

    Data bias

    "Historical and societal biases that are present in the data are used to train and fine-tune the model."

    From AI Risk Atlas (IBM2025)

  2. 65.19.01 · Risk Sub-Category

    Output risks (Fairness)

    Output bias

    "Generated content might unfairly represent certain groups or individuals."

    From AI Risk Atlas (IBM2025)

  3. 65.19.02 · Risk Sub-Category

    Output risks (Fairness)

    Decision bias

    "Decision bias occurs when one group is unfairly advantaged over another due to decisions of the model. This might be caused by biases in the data and also amplified as a result of the model’s training."

    From AI Risk Atlas (IBM2025)

  4. 65.15.04 · Risk Sub-Category

    Output risks (Value alignment)

    Toxic output

    "Toxic output occurs when the model produces hateful, abusive, and profane (HAP) or obscene content. This also includes behaviors like bullying."

    From AI Risk Atlas (IBM2025)

  5. 65.15.05 · Risk Sub-Category

    Output risks (Value alignment)

    Harmful output

    "A model might generate language that leads to physical harm The language might include overtly violent, covertly dangerous, or otherwise indirectly unsafe statements."

    From AI Risk Atlas (IBM2025)

  6. 65.23.04 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on affected communities

    "It is important to include the perspectives or concerns of communities that are affected by model outcomes when designing and building models. Failing to include these perspectives makes it difficult to understand the relevant context for the model and to engender trust within these communities."

    From AI Risk Atlas (IBM2025)

  7. 65.03.01 · Risk Sub-Category

    Training Data Risks (Privacy)

    Personal information in data

    "Inclusion or presence of personal identifiable information (PII) and sensitive personal information (SPI) in the data used for training or fine tuning the model might result in unwanted disclosure of that information."

    From AI Risk Atlas (IBM2025)

  8. 65.03.03 · Risk Sub-Category

    Training Data Risks (Privacy)

    Reidentification

    "Even with the removal or personal identifiable information (PII) and sensitive personal information (SPI) from data, it might be possible to identify persons due to correlations to other features available in the data."

    From AI Risk Atlas (IBM2025)

  9. 65.05.02 · Risk Sub-Category

    Training Data Risks (Intellectual property)

    Confidential information in data

    "Confidential information might be included as part of the data that is used to train or tune the model."

    From AI Risk Atlas (IBM2025)

  10. 65.11.03 · Risk Sub-Category

    Inference risks (Privacy)

    Personal information in prompt

    "Personal information or sensitive personal information that is included as a part of a prompt that is sent to the model."

    From AI Risk Atlas (IBM2025)

  11. 65.12.01 · Risk Sub-Category

    Inference risks (Intellectual property)

    Confidential data in prompt

    "Confidential information might be included as a part of the prompt that is sent to the model."

    From AI Risk Atlas (IBM2025)

  12. 65.12.02 · Risk Sub-Category

    Inference risks (Intellectual property)

    IP information in prompt

    "Copyrighted information or other intellectual property might be included as a part of the prompt that is sent to the model."

    From AI Risk Atlas (IBM2025)

  13. 65.16.02 · Risk Sub-Category

    Output risks (Intellectual Property)

    Revealing confidential information

    "When confidential information is used in training data, fine-tuning data, or as part of the prompt, models might reveal that data in the generated output. Revealing confidential information is a type of data leakage."

    From AI Risk Atlas (IBM2025)

  14. 65.20.01 · Risk Sub-Category

    Output risks (Privacy)

    Exposing personal information

    "When personal identifiable information (PII) or sensitive personal information (SPI) are used in training data, fine-tuning data, or as part of the prompt, models might reveal that data in the generated output. Revealing personal information is a type of data leakage."

    From AI Risk Atlas (IBM2025)

  15. 65.08.01 · Risk Sub-Category

    Training Data Risks (Robustness)

    Data poisoning

    "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."

    From AI Risk Atlas (IBM2025)

  16. 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)

  17. 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)

  18. 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)

  19. 65.09.04 · Risk Sub-Category

    Inference risks (Robustness)

    Prompt leaking

    "A prompt leak attack attempts to extract a model's system prompt (also known as the system message)."

    From AI Risk Atlas (IBM2025)

  20. 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)

  21. 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)

  22. 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)

  23. 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)

  24. 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)

  25. 65.18.01 · Risk Sub-Category

    Output risks (Robustness)

    Hallucination

    "Hallucinations generate factually inaccurate or untruthful content with respect to the model’s training data or input. This is also sometimes referred to lack of faithfulness or lack of groundedness."

    From AI Risk Atlas (IBM2025)

  26. 65.14.03 · Risk Sub-Category

    Output risks (misuse)

    Spreading toxicity

    "Generative AI models might be used intentionally to generate hateful, abusive, and profane (HAP) or obscene content."

    From AI Risk Atlas (IBM2025)

  27. 65.14.04 · Risk Sub-Category

    Output risks (misuse)

    Dangerous use

    "Generative AI models might be used with the sole intention of harming people."

    From AI Risk Atlas (IBM2025)

  28. 65.14.06 · Risk Sub-Category

    Output risks (misuse)

    Spreading disinformation

    "Generative AI models might be used to intentionally create misleading or false information to deceive or influence a targeted audience."

    From AI Risk Atlas (IBM2025)

  29. 65.14.05 · Risk Sub-Category

    Output risks (misuse)

    Nonconsensual use

    "Generative AI models might be intentionally used to imitate people through deepfakes by using video, images, audio, or other modalities without their consent."

    From AI Risk Atlas (IBM2025)

  30. 65.23.02 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on education: plagiarism

    "Easy access to high-quality generative models might result in students that use AI models to plagiarize existing work intentionally or unintentionally."

    From AI Risk Atlas (IBM2025)

  31. 65.23.05 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on education: bypassing learning

    "Easy access to high-quality generative models might result in students that use AI models to bypass the learning process."

    From AI Risk Atlas (IBM2025)

  32. 65.14.02 · Risk Sub-Category

    Output risks (misuse)

    Improper usage

    "Improper usage occurs when a model is used for a purpose that it was not originally designed for."

    From AI Risk Atlas (IBM2025)

  33. 65.15.03 · Risk Sub-Category

    Output risks (Value alignment)

    Over- or under-reliance

    "In AI-assisted decision-making tasks, reliance measures how much a person trusts (and potentially acts on) a model’s output. Over-reliance occurs when a person puts too much trust in a model, accepting a model’s output when the model’s output is likely incorrect. Under-reliance is the opposite, where the person doesn’t trust the model but should."

    From AI Risk Atlas (IBM2025)

  34. 65.23.08 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on human agency

    "AI might affect the individuals’ ability to make choices and act independently in their best interests."

    From AI Risk Atlas (IBM2025)

  35. 65.23.03 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on Jobs

    "Widespread adoption of foundation model-based AI systems might lead to people's job loss as their work is automated if they are not reskilled."

    From AI Risk Atlas (IBM2025)

  36. 65.23.07 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Human exploitation

    "When workers who train AI models such as ghost workers are not provided with adequate working conditions, fair compensation, and good health care benefits that also include mental health."

    From AI Risk Atlas (IBM2025)

  37. 65.16.01 · Risk Sub-Category

    Output risks (Intellectual Property)

    Copyright infringement

    "A model might generate content that is similar or identical to existing work protected by copyright or covered by open-source license agreement."

    From AI Risk Atlas (IBM2025)

  38. 65.21.03 · Risk Sub-Category

    Non-technical risks (legal compliance)

    Generated content ownership and IP

    "Legal uncertainty about the ownership and intellectual property rights of AI-generated content."

    From AI Risk Atlas (IBM2025)

  39. 65.23.01 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on cultural diversity

    "AI systems might overly represent certain cultures that result in a homogenization of culture and thoughts."

    From AI Risk Atlas (IBM2025)

  40. 65.01.01 · Risk Sub-Category

    Training Data Risks (Transparency)

    Lack of training data transparency

    "Without accurate documentation on how a model's data was collected, curated, and used to train a model, it might be harder to satisfactorily explain the behavior of the model with respect to the data."

    From AI Risk Atlas (IBM2025)

  41. 65.01.02 · Risk Sub-Category

    Training Data Risks (Transparency)

    Uncertain data provenance

    "Data provenance refers to tracing history of data, which includes its ownership, origin, and transformations. Without standardized and established methods for verifying where the data came from, there are no guarantees that the data is the same as the original source and has the correct usage terms."

    From AI Risk Atlas (IBM2025)

  42. 65.21.02 · Risk Sub-Category

    Non-technical risks (legal compliance)

    Legal accountability

    "Determining who is responsible for an AI model is challenging without good documentation and governance processes."

    From AI Risk Atlas (IBM2025)

  43. 65.22.01 · Risk Sub-Category

    Non-technical risks (Governance)

    Lack of system transparency

    "Insufficient documentation of the system that uses the model and the model’s purpose within the system in which it is used."

    From AI Risk Atlas (IBM2025)

  44. 65.22.02 · Risk Sub-Category

    Non-technical risks (Governance)

    Unrepresentative risk testing

    "Testing is unrepresentative when the test inputs are mismatched with the inputs that are expected during deployment."

    From AI Risk Atlas (IBM2025)

  45. 65.22.03 · Risk Sub-Category

    Non-technical risks (Governance)

    Incomplete usage definition

    "Since foundation models can be used for many purposes, a model’s intended use is important for defining the relevant risks of that model. As the use changes, the relevant risks might correspondingly change."

    From AI Risk Atlas (IBM2025)

  46. 65.22.04 · Risk Sub-Category

    Non-technical risks (Governance)

    Lack of data transparency

    "Lack of data transparency is due to insufficient documentation of training or tuning dataset details. "

    From AI Risk Atlas (IBM2025)

  47. 65.22.05 · Risk Sub-Category

    Non-technical risks (Governance)

    Incorrect risk testing

    "A metric selected to measure or track a risk is incorrectly selected, incompletely measuring the risk, or measuring the wrong risk for the given context."

    From AI Risk Atlas (IBM2025)

  48. 65.22.07 · Risk Sub-Category

    Non-technical risks (Governance)

    Lack of testing diversity

    "AI model risks are socio-technical, so their testing needs input from a broad set of disciplines and diverse testing practices."

    From AI Risk Atlas (IBM2025)

  49. 65.23.06 · Risk Sub-Category

    Non-technical risks (Societal impact)

    Impact on the environment

    "AI, and large generative models in particular, might produce increased carbon emissions and increase water usage for their training and operation."

    From AI Risk Atlas (IBM2025)

  50. 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)

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