AI Incident Database
Browse AI incidents
Every incident record (metadata only) from the weekly snapshot of 2026-09-07. Filter, then export the selection with its licence attached; each row links to the full record and its reports.
Filter incidents (1 active)
-
Amazon's AI Cameras Incorrectly Penalized Delivery Drivers for Mistakes They Did Not Make AIID ↗
Amazon's automated performance evaluation system involving AI-powered cameras incorrectly punished delivery drivers for non-existent mistakes, impacting their chances for bonuses and rewards.
-
Apple’s AVs Reportedly Struggled to Navigate Streets in Silicon Valley Test Drives AIID ↗
Apple’s autonomous cars were reported to have bumped into curbs and struggled to stay in their lanes after crossing intersections during an on-road test drives near the company’s Silicon Valley headquarters.
-
Epic Systems’s Sepsis Prediction Algorithms Revealed to Have High Error Rates on Seriously Ill Patients AIID ↗
Epic System's sepsis prediction algorithms was shown by investigators at the University of Michigan Hospital to have high rates of false positives and false negatives, allegedly delivering inaccurate and irrelevant information on patients, contrasting sharply with their published claims.
-
AI Tools Failed to Sufficiently Predict COVID Patients, Some Potentially Harmful AIID ↗
AI tools failed to sufficiently predict COVID patients, some potentially harmful.
-
Tesla's Autopilot Misidentified the Moon as Yellow Stop Light AIID ↗
Tesla's Autopilot was shown on video by its owner mistaking the moon for a yellow stop light, allegedly causing the vehicle to keep slowing down.
-
Facebook's Automated Moderation Flagged Gardening Group's Language Use by Mistake AIID ↗
Facebook's automated system flagged gardening groups' use of "hoe" and violent language against bugs as a violation by mistake.
-
Three Robots Collided, Sparking Fire in a Grocer's Warehouse in UK AIID ↗
A collision involving three robots at an Ocado's warehouse in Erith, UK, resulting in a fire but no reports of injuries.
-
SoftBank's Humanoid Robot, Pepper, Reportedly Frequently Made Errors, Prompting Dismissal AIID ↗
SoftBank's robot allegedly kept making mechanical errors, taking unplanned breaks, failing to recognize previously-met people, and breaking down during practice runs.
-
Algorithmic Allocation of Resources in Healthcare for Disabled and Elderly Care Services Allegedly Harmed Patients AIID ↗
A healthcare algorithm designed to equitably distribute caregiving resources is alleged to have drastically cut care hours for the disabled and elderly, purportedly leading to significant hardships and harm. Initially developed for fair resource allocation, the system reportedly faced legal challenges for its inability to accurately assess individual needs, resulting in allegedly reduced essential care.
-
Hiring Algorithms Provided Invalid Positive Results for Interview Responses in German AIID ↗
Two AI interview softwares provided positive but invalid results such as "competent" English proficiency and high match percentage for interview responses given in German by reporters.
-
Maryland Police Allegedly Relied on Facial Recognition Lead in Wrongful Arrest and Detention of Kimberlee Williams AIID ↗
Kimberlee Williams was reportedly arrested in Oklahoma on Maryland warrants after police allegedly relied on a facial recognition lead that incorrectly identified her as a suspect in bank fraud cases. The ACLU said Williams had never been to Maryland, police failed to investigate alibi evidence placing her in Oklahoma, and she spent about six months in jail before the charges were dropped.
-
Tesla FSD Misidentified Truck Hauling Traffic Lights as Trail of Traffic Lights AIID ↗
A Tesla driver posted on Twitter his Tesla FSD's "glitch," misidentifying deactivated traffic lights being carried by a truck as a constant trail of traffic lights while traveling at high speed on a highway.
-
McDonald's Reportedly Ends IBM Partnership After AI Drive-Thru Ordering Errors at U.S. Locations AIID ↗
Between 2021 and mid-2024, McDonald's reportedly piloted an AI-enabled voice ordering system that was developed with IBM and based on its 2019 acquisition of Apprente. It was reportedly deployed at at over 100 U.S. drive-thrus. Social media posts alleged frequent misorders, such as adding unwanted items, mixing adjacent lane orders, and ignoring corrections. In June 2024, McDonald's confirmed it ended the IBM pilot, citing plans to explore alternative voice-AI vendors.
-
Waymo Self-Driving Taxi Reportedly Behaved Unexpectedly, Driving away from Support Crew AIID ↗
A Waymo self-driving taxi car was reportedly shown on video stranded on a road in Arizona while carrying a passenger, suddenly driving away from the company's roadside assistance worker, and ended up being stuck farther down the road.
-
Tesla Phantom Braking Complaints Surged, Allegedly Linked to Tesla Vision Rollout AIID ↗
In late 2021, Tesla owners’ complaints to the National Highway Traffic Safety Administration about sudden unexpected automatic braking rapidly increased, coinciding with when radar was no longer equipped in its Model 3 and Model Y vehicles.
-
Betfair's Machine-Learning Risk System Reportedly Failed to Flag Luke Ashton Before Gambling-Related Suicide in England AIID ↗
In England, Luke Ashton died by suicide on April 22, 2021 after a longstanding gambling disorder. A 2023 inquest and subsequent Prevention of Future Deaths report found that Betfair had assessed him as a low-risk gambler, did not meaningfully intervene between 2019 and his death, and used an algorithm that failed to flag his escalating gambling activity even as his betting intensified and his financial exposure grew.
-
Algorithmic Staffing Failures Linked to Resident Deaths at Leading Assisted-Living Chain Brookdale AIID ↗
Brookdale Senior Living's algorithm-based staffing system, "Service Alignment," reportedly left facilities understaffed, leading to critical incidents. For example, on April 21, 2021, Louise Walker, a resident at Brookdale's Jacksonville facility, died after falling and being left unattended for over two hours. State investigators cited Brookdale for medical neglect. The algorithm has been linked to multiple incidents of neglect, injuries, and deaths, prompting lawsuits.
-
Auto-Insurance Photo-Based Estimation Allegedly Gave Inaccurate Repair Prices Frequently AIID ↗
Auto-insurance companies' photo-based estimation of repair price was alleged by repair shop owners and industry groups as providing inaccurate estimates, causing damaged cars to stay in the shop longer.
-
Tesla on Autopilot Crashed Parked Michigan Police Car on Interstate AIID ↗
A Tesla Model Y on Autopilot collided with a parked Michigan State Police (MSP) car which had its emergency lights on, in Eaton County, Michigan, although no one was injured.
-
How French welfare services are creating ‘robo-debt’ AIID ↗
A French welfare office using software to automatically evaluate cases incorrectly notified a woman receiving benefits that she owed €542.
-
Dartmouth's Geisel School of Medicine Reportedly Used Canvas Activity Logs to Accuse Students of Cheating AIID ↗
Dartmouth's Geisel School of Medicine reportedly used Canvas learning-management activity logs and an internal analysis process to investigate remote-exam cheating during the 2020–2021 academic year. Seventeen medical students were charged after the school inferred that they accessed course materials during exams, but students and outside technical reviewers said automated Canvas activity may have been misread as intentional misconduct. Dartmouth later dropped the charges.
-
Year-long AI Surveillance Pilot in Two South Australian Aged Care Facilities Reportedly Overwhelmed Staff with False Positives AIID ↗
Between March 2021 and March 2022, an AI-enabled video and audio monitoring system was trialed in two South Australian aged care facilities. According to an independent audit commissioned by South Australia Health, the system produced over 12,000 false alerts, overwhelming staff and contributing to at least one missed real incident. The report concluded the system did not reach accuracy levels acceptable to staff or management during the pilot.
-
Facebook's Automated Tools Failed to Adequately Remove Hate Speech, Violence, and Incitement AIID ↗
Facebook's automated moderation tools were shown by internal documents performing incomparably to human moderators, and accounting for only a small fraction of hate speech, violence, and incitement content removal.
-
Manufacturing Robot Failure Caused Factory Worker's Death in India AIID ↗
A sensor snag resulted in an automotive parts factory robot falling on a factory worker in India
-
Facebook’s and Twitter's Automated Content Moderation Reportedly Failed to Effectively Enforce Violation Rules for Small Language Groups AIID ↗
Facebook's and Twitter were not able to sufficiently moderate content of small language groups such as the Balkan languages using AI, allegedly due to the lack of investment in human moderation and difficulty in AI-solution design for the languages.
Disclaimer: informational only, not legal advice. Verify every claim against the linked official sources and consult a qualified lawyer before acting.
Frequently asked questions
What is recorded in each incident?
What happened, when, who was involved as developer or deployer, who was harmed, the sectors and countries affected, and the domain and subdomain the harm falls under. Incidents are mirrored from the AI Incident Database and each links back to its source record.
Does an incident mean wrongdoing was proven?
No. These are reported incidents, not findings of liability. Entities named in a report are named as reported, allegations are marked as alleged, and nothing here is a legal determination.
How current is the incident data?
It syncs directly from the AI Incident Database several times a day, so a new incident usually appears within hours. Each record carries the date it was last synced, and the snapshot date is shown on the page.
Does this page include the full incident reports?
No. The report texts are outside the licence this data is shared under, so only the metadata and the classifications appear here. Every row links to the original incident record, which carries the reports themselves.