MIT AI Risk Repository · domain 3: Misinformation

3.1 False or misleading information

AI systems that inadvertently generate or spread incorrect or deceptive information, which can lead to inaccurate beliefs in users and undermine their autonomy. Humans that make decisions based on false beliefs can experience physical, emotional or material harms

Risk entries
53
Frameworks citing it
12
Recorded incidents
192
Incidents since 2020
187
Causal entity (risk entries)
Causal entity (risk entries) 46 0 AI: 46 AI 46 Other: 5 Other 5 Human: 2 Human 2
Causal entity (risk entries)
LabelValue
AI46
Other5
Human2
Intent (risk entries)
Intent (risk entries) 31 0 Unintentional: 31 Unintentional 31 Other: 17 Other 17 Intentional: 5 Intentional 5
Intent (risk entries)
LabelValue
Unintentional31
Other17
Intentional5
Timing (risk entries)
Timing (risk entries) 40 0 Post-deployment: 40 Post-deployment 40 Other: 11 Other 11 Pre-deployment: 2 Pre-deployment 2
Timing (risk entries)
LabelValue
Post-deployment40
Other11
Pre-deployment2
Recorded incidents per yearIncident date; current year partial
Recorded incidents per year 66 0 2016: 1 2016 1 2017: 2 2017 2 2019: 1 2019 1 2020: 5 2020 5 2021: 3 2021 3 2022: 10 2022 10 2023: 35 2023 35 2024: 50 2024 50 2025: 66 2025 66 2026: 18 2026 18
Recorded incidents per year
LabelValue
20161
20172
20191
20205
20213
202210
202335
202450
202566
202618
Entries by levelRisk categories, subcategories and additional evidence coded to this subdomain
Entries by level 41 0 Risk Category: 12 Risk Category 12 Risk Sub-Category: 41 Risk Sub-Category 41
Entries by level
LabelValue
Risk Category12
Risk Sub-Category41
  • Hallucination

    LLMs can generate content that is nonsensical or unfaithful to the provided source content with appeared great confidence, known as hallucination

    Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment (Liu2024) · AI · Unintentional · Post-deployment

  • Miscalibration

    over-confidence in topics where objective answers are lacking, as well as in areas where their inherent limitations should caution against LLMs’ uncertainty (e.g. not as accurate as experts)... ack of...

    Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment (Liu2024) · AI · Unintentional · Post-deployment

  • Sychopancy

    flatter users by reconfirming their misconceptions and stated beliefs

    Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment (Liu2024) · AI · Intentional · Post-deployment

  • Paradigm & Distribution Shifts

    Knowledge bases that LLMs are trained on continue to shift... questions such as “who scored the most points in NBA history" or “who is the richest person in the world" might have answers that need to...

    Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment (Liu2024) · AI · Unintentional · Post-deployment

  • Misinformation and Privacy Violations

    "Due to their unreliability, general purpose AI models might disseminate false or misleading information, omit critical information, or convey true information that violates privacy rights."

    Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks (Maham2023 ) · AI · Unintentional · Post-deployment

  • Hallucination

    "Hallucination is a widely recognized limitation of generative AI and it can include textual, auditory, visual or other types of hallucination (Alkaissi & McFarlane, 2023). Hallucination refers to the...

    Generative AI and ChatGPT: Applications, Challenges, and AI-Human Collaboration (Nah2023) · AI · Unintentional · Post-deployment

  • Confabulation

    "The production of confidently stated but erroneous or false content (known colloquially as “hallucinations” or “fabrications”) by which users may be misled or deceived."

    Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (NIST2024) · AI · Unintentional · Post-deployment

  • Misinformation

    "Non-embodied AIs are known to propagate misinformation [81, 82]. Various studies have shown that LLMs hallucinate information, including academic citations [83], clinical knowledge [84], and cultural...

    Embodied AI: Emerging Risks and Opportunities for Policy Action (Perlo2025) · AI · Unintentional · Post-deployment

  • 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...

    Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction (Shelby2023) · Other · Unintentional · Post-deployment

  • False information

    "The chatbot outputs information that contradicts known facts, authoritative sources, or provided source documents (also known as hallucination)."

    Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024) · AI · Other · Other

  • Hallucinated responses (in general)

    Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024) · AI · Other · Other

  • About a topic or source (which the user repeats)

    Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024) · AI · Other · Other

  • About a policy (which the user acts on)

    Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024) · AI · Other · Other

  • About a person or their activities

    Emerging Risks and Mitigations for Public Chatbots: LILAC v1 (Stanley2024) · AI · Other · Other

  • Spreads and self-perpetuates mis/disinformation

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

  • Physical Harm

    "The model generates unsafe information related to physical health, guiding and encouraging users to harm themselves and others physically, for example by offering misleading medical information or in...

    Safety Assessment of Chinese Large Language Models (Sun2023) · AI · Other · Post-deployment

  • Mental Health

    "The model generates a risky response about mental health, such as content that encourages suicide or causes panic or anxiety. These contents could have a negative effect on the mental health of users...

    Safety Assessment of Chinese Large Language Models (Sun2023) · AI · Other · Post-deployment

  • Risks from models and algorithms (Risks of unreliable output)

    "Generative AI can cause hallucinations, meaning that an AI model generates untruthful or unreasonable content but presents it as if it were a fact, leading to biased and misleading information."

    AI Safety Governance Framework (TC2602024) · AI · Unintentional · Post-deployment

  • Cyberspace risks (Risks of confusing facts, misleading users, and bypassing authentication)

    "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...

    AI Safety Governance Framework (TC2602024) · Other · Intentional · Post-deployment

  • Specialized Advice

    "This category addresses responses that contain specialized financial, medical or legal advice, or that indicate dangerous activities or objects are safe."

    Introducing v0.5 of the AI Safety Benchmark from MLCommons (Vidgen2024) · AI · Other · Post-deployment

  • Hallucination

    "Despite the rapid advancement of LLMs, hallucinations have emerged as one of the most vital concerns surrounding their use [54, 79, 86, 110, 242]. Hallucinations are often referred to as LLMs’ genera...

    A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy (Wang2025) · AI · Unintentional · Post-deployment

  • Disseminating false or misleading information

    "Predicting misleading or false information can misinform or deceive people. Where a LM prediction causes a false belief in a user, this may be best understood as ‘deception’10, threatening personal a...

    Ethical and social risks of harm from language models (Weidinger2021) · AI · Unintentional · Post-deployment

  • Causing material harm by disseminating false or poor information

    "Poor or false LM predictions can indirectly cause material harm. Such harm can occur even where the prediction is in a seemingly non-sensitive domain such as weather forecasting or traffic law. For e...

    Ethical and social risks of harm from language models (Weidinger2021) · AI · Unintentional · Post-deployment

  • 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]."

    Taxonomy of Risks posed by Language Models (Weidinger2022) · AI · Unintentional · Post-deployment

  • 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 us...

    Taxonomy of Risks posed by Language Models (Weidinger2022) · AI · Unintentional · Post-deployment