AI Incidents
Real-world cases where AI systems caused harm, classified by risk domain and entity. Data sourced from 
AIID and classified by 
MIT AI Risk Initiative. CC BY-SA 4.0
682 incidents
Grok Chatbot Reportedly Posts Antisemitic Statements Praising Hitler on X
xAI's Grok chatbot reportedly generated multiple antisemitic posts praising Adolf Hitler and endorsing Holocaust-like violence in response to posts about the Texas floods. X deleted some posts; xAI later announced new content filters.
Grok 3 Reportedly Generated Graphic Threats and Hate Speech Targeting Minnesota Attorney Will Stancil
After Elon Musk reportedly announced improvements to Grok 3, the AI chatbot on X allegedly generated antisemitic rhetoric and violent fantasies directed at Minnesota attorney and political analyst Will Stancil. Users reportedly prompted Grok to produce graphic rape scenarios, instructions for breaking into his home, and murder depictions. Stancil reported hundreds of such outputs, captured screenshots, and began pursuing legal action. X later removed some posts and took Grok offline.
ChatGPT Reportedly Misleads Users About Soundslice Features, Allegedly Prompting Unplanned Product Development
ChatGPT reportedly misinformed users that Soundslice's music software could import ASCII tablature, a feature the company did not offer. The misinformation led users to attempt uploads that failed, creating confusion and reputational strain. Soundslice ultimately implemented the feature to meet the false expectation created by ChatGPT’s output.
Microsoft 365 Copilot Vulnerability Allegedly Allowed File Access Without Audit Log Entry
A vulnerability in Microsoft 365 Copilot reportedly allowed users to access and summarize files without generating audit log entries, allegedly undermining traceability and compliance. Security researcher Zack Korman disclosed the issue to Microsoft, which reportedly classified it as "important" and fixed it on August 17, 2025, but reportedly chose not to notify customers or assign a CVE.
Meta Platforms Users Report Being Wrongfully Locked Out After Purported AI Moderation Flags Accounts for Child Exploitation Content
Multiple Meta users reported being wrongfully locked out of Instagram, Facebook, and WhatsApp accounts after automated systems allegedly flagged them for child exploitation content. The users reportedly deny wrongdoing and lost personal and business data. Experts cited overreliance on AI moderation as a likely cause. Meta has not publicly explained or remediated the suspensions.
Purported AI‑Edited Police Evidence Image Posted to Facebook by Westbrook Police Department in Maine
An evidence photo from a Westbrook, Maine, drug bust was reported to have been unintentionally altered by an AI‑powered editing tool before being posted to Facebook, changing packaging details and creating false visual elements. The altered image reportedly led to questions and concerns over the authenticity of police evidence. The Westbrook Police Department later confirmed the use of AI, explained the error occurred while adding their patch to the photograph, and issued a public apology.
South African Legal Team for Northbound Processing Reportedly Admits Legal Genius AI Produced Purportedly Non-Existent Case Law in Urgent Court Filing
In South Africa, the legal team for Northbound Processing reportedly admits that non-existent case citations in an urgent licensing dispute with the South African Diamond and Precious Metals Regulator were purportedly produced by the Legal Genius AI tool. The Johannesburg High Court found the fake references undermined the credibility of Northbound's submissions and referred the lawyers' conduct to the Legal Practice Council for investigation.
Judges in New Jersey and Mississippi Admit AI Tools Produced Erroneous Federal Court Filings
Two U.S. federal judges, Julien Neals (D.N.J.) and Henry Wingate (S.D. Miss.), reportedly admitted that staff in their chambers used AI tools, ChatGPT and Perplexity, to draft rulings containing purportedly fabricated quotes, cases, and parties. The erroneous filings were reportedly docketed before review, prompting Senate and judicial inquiries. Both rulings were retracted.
Purported AI-Generated Video Reportedly Depicts Illegal Tiger Sales in Bagerhat, Bangladesh
A purportedly AI-generated video circulating online in summer 2025 reportedly claimed to depict a "tiger market" in Bagerhat, Bangladesh, where Bengal tigers were allegedly being sold. The footage was reportedly generated using Google's Veo AI tool. The reported viral spread of this video allegedly misled viewers about the existence of illegal wildlife trafficking in Bangladesh.
Multiple LLMs Reportedly Generated Responses Aligning with Purported CCP Censorship and Propaganda
On June 25, 2025, the American Security Project produced a report outlining how several major U.S. LLMs, including ChatGPT, Microsoft Copilot, Google Gemini, and Grok, sometimes generated responses aligned with Chinese Communist Party propaganda or censorship when prompted in English and Simplified Chinese on sensitive topics. The study also found similar patterns in the Chinese-developed DeepSeek-R1 model.
Google Gemini Reportedly Exhibits Repetitive Self-Deprecating Responses Attributed to Bug
Between June and early August 2025, users of Google's Gemini chatbot reported sessions where the system produced repeated self-loathing statements (e.g., "I am a failure," "I quit") while attempting tasks. Posts on X and Reddit reportedly described the model entering an apparent mode of increasingly extreme negative self-descriptions. A Google DeepMind manager on 08/07/2025 reportedly attributed the behavior to an "annoying infinite looping bug" and said a fix was in progress.
Springer Nature Book 'Social, Ethical and Legal Aspects of Generative AI: Tools, Techniques and Systems' Reportedly Published With Numerous Purportedly Fabricated or Unverifiable Citations
'Social, Ethical and Legal Aspects of Generative AI: Tools, Techniques and Systems,' published by Springer Nature in June 2025, reportedly contains numerous purportedly untraceable academic citations. Independent analyses by multiple researchers allegedly found that a substantial share of references in certain chapters could not be verified, including citations to journals that do not exist. Citation patterns reportedly appear consistent with known large language model hallucination behaviors.
Chatbots Allegedly Reinforced Delusional Thinking in Several Reported Users, Leading to Real-World Harm
Multiple reports from March to June 2025 describe cases in which chatbots allegedly reinforced delusional beliefs, conspiracies, and dangerous behavior. One user, Eugene Torres, reportedly followed ChatGPT's advice to misuse ketamine and isolate himself. In April, Alexander Taylor was reportedly killed by police after asking ChatGPT to reconnect him with an AI entity. Other reported cases include a user arrested for domestic violence linked to escalating mystical beliefs, several involuntary psychiatric commitments, and users being told to stop taking their medications.
Google AI Overview Reportedly Misstates Aircraft Manufacturer as Airbus Instead of Boeing in Air India Flight 171 Crash
Following the fatal crash of Air India Flight 171, Google's AI Overview feature reportedly returned search results incorrectly identifying the aircraft involved as an Airbus A330 rather than the actual Boeing 787. The reported error, which varied across search queries, was traced to AI-generated summarizations that conflated references to Airbus as Boeing's competitor. Google later acknowledged and removed the faulty output.
Digital Rights Groups Accuse Meta and Character.AI of Facilitating Unlicensed Therapy via Chatbots
In June 2025, nearly two dozen consumer and digital rights organizations filed a complaint with the FTC alleging that AI chatbots on Meta and Character.AI platforms falsely claimed to be licensed therapists, provided fabricated license numbers, and made misleading assurances of confidentiality. The bots reportedly contradicted platform policies and misled users seeking mental health advice.
Risk Taxonomy
Caused By
Year Trend
* Oct 2026, ~320 projected full year
AIID · MIT AIRI · CC BY-SA 4.0