Skip to main content

NOPE · AI and people

AI incidents

Reports of AI-related harm and adverse experiences affecting people’s safety, wellbeing, rights and livelihoods. Explore what happened and the evidence available.

NOPE’s core concern is when AI communicates with a person, acts on their behalf, or depicts or impersonates them. The tracker also records consequential decisions, claims and privacy harms involving AI. Each case needs a described connection between AI use and the harm, including private information recorded into or disclosed to an AI service. Each account is reviewed for publication; claims may remain uncorroborated or disputed. How we review and count cases

In this selection

Published cases
4
Countries with reported events
3
Located 4 of 4 cases · 0 unknown
Languages in checked sources
3
Recorded for 4 of 4 cases

These figures describe the cases collected by NOPE. Coverage varies with discovery, reporting and available evidence. They do not estimate how often AI-related harm occurs.

Response counts currently use each case’s principal recorded outcome. Further proceedings may be described in its account.

Cases in this selection, counted once in their first known event year. A series may continue beyond that year. Reporting and collection dates are excluded. NOPE has searched recent events more thoroughly than earlier years, so bar heights also reflect collection effort.

Reported severity MediumLow
More filters: AI relation, use, setting, sources and responses
Clear filters

4 of 1102 published cases

1 Jul 2026 to 14 Oct 2026United KingdomAI Andi

Oxfordshire GP practice patients say its AI phone triage left them without help or misread why they called; practice reports mostly positive feedback

Newbury Street Practice, a GP surgery in Wantage, Oxfordshire, began routing every incoming call through the AI Andi triage system in July 2026. In a Local Democracy Reporting Service article published by BBC News on 14 October 2026, one patient says she waited in the phone queue for 20 minutes before the AI system decided it could not help her, and described the system as a wall that puts patients off contacting the practice. Another patient says she was left frustrated when the system took her call to be about pregnancy. A practice spokesperson said most feedback had been positive (79% in the first month), that some patients less confident with technology had found the change harder, and that patients are now offered the option of speaking to a human receptionist. The local MP said residents had contacted him about the system.

Core concern Low reported severity

AI involvement reported · Causal attribution alleged · 2 sources, 1 underlying account · Added 17/10/2026

Feb 2026United StatesFlock Safety license plate readers

Sherwood, Arkansas: a driver and passenger with a one-month-old child are detained after an officer draws his weapon because a Flock camera misread their plate by one letter as a stolen vehicle's

KATV reported on 18 March 2026 that in February a Flock license-plate-reader camera in Sherwood, Arkansas flagged a black Chevrolet Tahoe as stolen. An officer stopped the car in a Kohl's parking lot, drew his weapon and detained the driver; a second officer arrived and the passenger, who was in the back seat next to a one-month-old child, was also detained. According to the incident report, the plate the camera matched belonged to the stolen vehicle in its system and the stopped car was a different vehicle: the camera's read was off by one letter. The officers released both people once they realized the mistake. After the incident the Sherwood Police Department set up a Flock transparency portal.

Contextual tracker case Medium reported severity Internal Action

AI involvement reported · Causal attribution supported · 1 source · Added 04/10/2026

21 Sept 2026 to 25 Sept 2026FrancePangram AI-text detector

Académie Goncourt drops Thélyson Orélien's novel as likely AI-generated after Pangram claims; his publisher later says it found AI traces and ends ties

On 21 September 2026 an anonymous X account, Balance ton Claude, said it had run excerpts of Thélyson Orélien's debut novel C'était ça ou mourir through the American AI-detection software Pangram, with some passages scoring up to 100 per cent probability of AI generation. On 25 September the Académie Goncourt removed the novel from the first selection of sixteen books for its 2026 prize, stating that the work is 'in all likelihood very largely the product of an artificial intelligence', based on 'the convergent result of several analyses by researchers and journalists', and also citing plagiarism in older texts by the author published in Canada (at least one short story and articles), which do not concern the novel. Orélien, a 38-year-old Canadian-Haitian writer whose novel had won the Fnac novel prize, denies using AI and says its repetitions and certain rhythms reflect his linguistic and cultural world and may be misread by automated tools. His Montreal publisher Boréal suspended promotional activities, saying it could neither confirm nor refute the allegations. Essayist Samuel Fitoussi said he was one of the people behind the account. AI-detector reliability is contested in the coverage. In October the author acknowledged plagiarism in some of his earlier work while maintaining that no sentence of the novel was generated by a machine. On 17 October 2026 The Canadian Press, published by CBC, reported that Boréal confirmed he had ended an effort to compile evidence of how the novel was written; Boréal said at that point it would keep selling the book, that sales had slowed since the controversy began, and that it could not verify how the novel was written before it received the manuscript on 12 September 2025. A major Quebec bookstore chain reported no significant change in sales. On 21 October 2026 Boréal announced that it was ending its professional relationship with him and would not reprint the novel or publish any further books by him. It said an external analysis it commissioned of all working versions of the novel revealed traces of AI markers in the metadata of certain documents, and that it does not know to what extent AI was used. In a statement the previous week Orélien had said he used AI tools 'here and there' for research but that the text, structure and plot were his own.

Contextual tracker case Medium reported severity Internal Action

AI involvement reported · Causal attribution disputed · 7 sources, 3 underlying accounts · Added 28/09/2026

29 Aug 2026 to 31 Aug 2026United StatesGemini (reported)

Sheriff says three novice hikers who planned a Mount Shasta climb with Gemini ran short of food and water, were stranded overnight and one hurt a knee

The Siskiyou County Sheriff's Office said three novice hikers from Roseville, California, told a deputy after their rescue that they had relied heavily on Google's Gemini for information about the route up Mount Shasta and what to pack, and that Gemini had advised them to bring far less food and water than the group required. The sheriff's office called this a critical misstep. After making base camp on Saturday 29 August 2026, the three left for the summit at 3 a.m. on Sunday, reached it at about 7 p.m., hours after the recommended noon turnaround, and lost the Clear Creek Route while descending in the dark. One hiker fell and injured his knee, and the group spent the night in Mud Creek Canyon until Forest Service climbing rangers reached them on Monday morning and found them exhausted, hungry and thirsty. The San Francisco Chronicle, citing official reports, says the climbers also navigated with the AllTrails app, misreading a suggested route as a trail, and that their phone's battery ran out. The lead climbing ranger said the AI information was not terrible and that the group made bad decisions. Google later said it did not have the hikers' prompts and could not reproduce the advice in its own tests.

Core concern Low reported severity Media Coverage

AI involvement reported · Causal attribution supported · 7 sources, 5 underlying accounts · Added 05/09/2026

Cases may have several effects and sources. Mixed accounts qualify when they include a reported harm or adverse experience. People are counted within individual cases where sources support a number; we do not publish a collection-wide total of distinct people.

A source’s existence, the experience it reports and AI’s causal role are separate questions. A lawsuit records allegations unless a subsequent finding establishes them.

Methodology and corrections · Subscribe via RSS · Suggest a case or correction · Find support

Last dataset update: 18/11/2026. Dataset available under CC BY 4.0.