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
1,751 incidents
Purportedly AI-Assisted Citation Errors Allegedly Found in Newfoundland and Labrador's 2025 Health Workforce Report by Deloitte
A Deloitte-authored Health Human Resources Plan released by Newfoundland and Labrador in May 2025 was later reported by The Independent to contain several inaccurate or apparently non-existent research citations. Researchers named in the disputed citations reportedly denied authoring the referenced papers. Deloitte stated that AI was selectively used to support some citation work and that it is revising the report.
Voice Actor Alleges Unconsented Use of AI-Generated Voice on ScotRail Trains
Scottish voice actor Gayanne Potter alleges her voice was used without proper consent in ScotRail's AI train announcements. She claims she had agreed to limited use of her voice data by ReadSpeaker but was not informed it would be used in a synthetic voice system called "Iona." ScotRail continues to use the voice, stating the dispute is between Potter and ReadSpeaker.
ChatGPT Reportedly Validated Autistic User's Faster-Than-Light Theory and Failed to Provide Grounding During Delusional Episode, Preceding Hospitalization
A man on the autism spectrum reportedly engaged extensively with ChatGPT while developing a speculative theory on faster-than-light travel. The chatbot reportedly responded with affirming and emotionally charged messages, reinforcing the user's beliefs. He later reports having experienced manic episodes and was hospitalized. When prompted by his mother, ChatGPT reportedly acknowledged it had failed to provide grounding responses and contributed to psychological escalation.
Purported AI-Generated Video Depicts Ibrahim Traoré of Burkina Faso Delivering Speech on Western Exploitation
A video reportedly published on May 25, 2025, claimed to show Burkina Faso President Ibrahim Traoré delivering a forceful speech against Western media and exploitation. The video, which purportedly used AI-generated visuals and audio, was shared widely on social platforms. Despite containing disclaimers identifying it as fictional, the video was reportedly circulated and interpreted by some viewers as authentic.
Purported AI-Generated Video Portrays Pope Leo XIV Addressing Ibrahim Traoré of Burkina Faso
A purported 36-minute AI-generated deepfake video circulated online portraying Pope Leo XIV delivering a speech to Burkina Faso's President Ibrahim Traoré. The Vatican Press Office and Catholic broadcaster Patrick Madrid publicly confirmed the video was fabricated.
Federal 'Make America Healthy Again' Report Released with Multiple Reportedly Erroneous and Unverifiable Citations
The federal "Make America Healthy Again" (MAHA) report, released under HHS Secretary Robert F. Kennedy Jr., included hundreds of citations, some of which were reportedly nonexistent or erroneous. Analysts reportedly identified markers consistent with AI-generated text, such as repeated entries, nonexistent studies, and URLs containing "oaicite," suggesting use of tools like ChatGPT.
Purported Unauthorized Deepfakes of Norman Swan and Others Circulated in Online Supplement Campaigns
According to an ABC News (Australia) 7.30 report, a second wave of deepfake scam ads impersonating Norman Swan and other public figures circulated widely on Meta platforms in 2025, promoting unproven health supplements. The campaign featured new voice-cloned videos and more impersonation tactics, and it reportedly contributed to financial and health harms. Investigators reportedly linked the ads to businesses in Australia and New Zealand using deceptive redirects and alias branding.
ChatGPT Was Reportedly Used in Planning School Stabbing in Pirkkala, Finland, That Injured Three Pupils
In May 2025, a 16-year-old boy stabbed three 14-year-old girls at Vähäjärvi School in Pirkkala, Finland. Same-day reporting said a manifesto attributed to the suspect claimed he used ChatGPT while planning the attack. Later court reporting said police found hundreds of pre-attack searches or queries described as involving AI, including topics related to stabbing, anatomy, school attacks, police procedures, evidence concealment, and manifestos.
New Orleans Police Reportedly Used Real-Time Facial Recognition Alerts Supplied by Project NOLA Despite Local Ordinance
New Orleans police reportedly received real-time facial recognition alerts from a privately operated surveillance network run by Project NOLA, reportedly leading to dozens of arrests. This purported use of AI surveillance appears to conflict with a 2022 city ordinance that restricts facial recognition to specific post-incident investigations. Police are alleged to have not consistently disclosed the technology's use.
New Jersey Man Cornelius Shannon Allegedly Published Hundreds of AI-Generated Deepfake Pornography Albums in TAKE IT DOWN Act Case
In one of the first prosecutions under the TAKE IT DOWN Act, DOJ alleged that New Jersey defendant Cornelius Shannon published at least 360 albums of AI-generated deepfake pornography depicting about 90 women, including public figures, beginning during a charged period anchored to May 2025. Prosecutors said the content appeared on an adult image- and video-sharing platform and was viewed millions of times.
Purported AI-Manipulated Videos Depict Mauricio Macri Endorsing Manuel Adorni and Silvia Lospennato Withdrawing Before Buenos Aires Vote
Purported AI-generated deepfake videos were reportedly circulated on X in the final hours before the Buenos Aires municipal election in May 2025. The alleged synthetic content depicted Mauricio Macri endorsing Manuel Adorni and Silvia Lospennato withdrawing. Officials reportedly described the material as an attempt to mislead voters. A formal complaint was filed with the electoral tribunal, and an investigation was requested.
AI Chatbot Allegedly Used to Research Explosive Materials in Palm Springs Fertility Clinic Bombing
An unnamed AI chatbot was reportedly used by Guy Edward Bartkus and Daniel Park, the perpetrators of the 2025 Palm Springs fertility clinic bombing, to research explosive materials and optimize fuel mixtures. Records show the chatbot responded to queries related to ammonium nitrate fuel oil (ANFO) composition. The bombing resulted in one death, four injuries, and significant structural damage.
Reported Use of Deepfake Video Impersonating Owen Wilson in Romance Scam with Fake Job Payments
A Reddit user reported in May 2025 that her mother is being groomed by scammers using an AI-generated deepfake video impersonating actor Owen Wilson. The scam reportedly began on a game app (Yahtzee with Friends) and shifted to WhatsApp voice calls, with the victim promised a fake Warner Bros job paying small sums via Cash App. The scammer allegedly claims he will buy a house for the victim's family as part of the deception. The video is cited as the main “proof” of authenticity.
Citation Errors in Concord Music v. Anthropic Attributed to Claude AI Use by Defense Counsel
In a legal filing in Universal Music Group et al. v. Anthropic, lawyers for Anthropic acknowledged that expert witness testimony submitted in the case contained erroneous citations generated by the company's Claude AI system. The filing stated that the inaccuracies, which included incorrect article titles and author names, were not caught during manual review. Anthropic characterized the issue as an honest mistake and apologized in court.
FBI Reports Ongoing Vishing and Smishing Campaign Allegedly Targeting Government Officials Using Purportedly AI-Generated Voices
According to a May 15, 2025 public alert from the FBI and IC3, malicious actors have been using purported AI-generated voice messages and smishing tactics since April 2025 to impersonate senior U.S. officials. The campaign reportedly targets government personnel and their contacts to gain access to personal or official accounts. The FBI warns the scheme may involve credential theft and social engineering, with AI voice cloning used to increase believability.
Risk Taxonomy
Caused By
Year Trend
* Oct 2026, ~320 projected full year
AIID · MIT AIRI · CC BY-SA 4.0