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 is wider: it also records consequential decisions and claims about people. Each account is reviewed for publication; claims may remain uncorroborated or disputed. How we review and count cases

In this selection

Published cases
5
Countries with reported events
3
Located 4 of 5 cases · 1 unknown
Languages in checked sources
4
Recorded for 5 of 5 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.

No known event years in this selection.

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

5 of 174 published cases

Event date unknownRomaniaLia, Libra Internet Bank's generative-AI customer assistant (underlying model not reported)

Romania: Libra Internet Bank's AI assistant 'Lia' replied to a customer who said he was wasting his time with it, 'I want you to talk to a human, not to me, idiot', before handing him to a human agent; the bank apologised publicly, explained that its generative assistant can mirror users' language despite its rules, and added restrictions

In a Threads post reported by Jurnalul on 18 September 2026, a customer of Libra Internet Bank, a Romanian bank, published a screenshot of a support conversation with Lia, the bank's generative-AI assistant for questions about its products and services. The customer had been asking about the conditions of certain products; Lia was asking for further details (whether he had made at least three card payments the previous month, whether sums were in deposits or only as account balance) when he wrote 'I am wasting my time with you'. Lia replied, in Romanian: 'I want you to talk to a human, not to me, idiot. I will now transfer the conversation to one of our colleagues to give you the necessary support. You will receive an answer shortly.' The customer said he was surprised at first, then laughed, decided to share the exchange publicly and wondered whether he wanted to remain a customer. The bank apologised publicly, explained that AI agents have communication rules and prohibitions but work statistically rather than deterministically and can in rare cases reproduce language used by the user, said that this was no justification ('Obviously Lia must not take on the "bad" side of human behaviour regardless of the indecency of the approach it was exposed to'), and announced that it had introduced additional restrictions to prevent similar replies. Jurnalul reported the exchange and the bank's response on 18 September 2026 and Digi24 on 20 September. The customer is not named.

Core concern News report AI involvement: supported Causal attribution: established Low reported severity Internal Action

Added 22/09/2026

Event date unknownBrazilAI voice cloning (tool not identified)

Campo Grande, Mato Grosso do Sul: a workshop owner paid about R$28,000 by Pix for a car after messages and voice notes in his brother's AI-cloned voice 'confirmed' the vehicle; a second family avoided the same fraud by checking directly with their daughter

A businessman who owns a workshop in Campo Grande, Mato Grosso do Sul, found a car advertised at a market price by a seller who said he was in Ponta Porã, the businessman's home town; he passed his brother's contact so the brother could inspect the car. The fraudsters then contacted him as if they were his brother, saying he had seen the car, sending a video and confirming everything was in order; believing he was speaking to his brother, he made a Pix transfer of about R$28,000, after which the contact was blocked. He told G1 MS and TV Morena that the criminals used information about his family and a cloned voice. Guided by his wife, he contested the transfer through the Central Bank's special return mechanism and, after nearly three hours, a police report and an in-person registration, recovered about 20 percent. In a second case reported alongside, a medical student's mother in Dourados received messages, a voice note and a video from someone posing as her daughter, in a cloned voice, approving payment for an HB20 she was buying in Campo Grande; the mother checked directly with her daughter and discovered the audio had been AI-generated before any money was sent. A state prosecutor said his unit had not yet catalogued identical voice-cloning cases and urged victims to file police reports; the Civil Police did not respond to G1's questions. The date of the fraud is not reported.

Core concern News report AI involvement: reported Causal attribution: alleged Medium reported severity Media Coverage

Added 21/09/2026

Event date unknownUnited StatesFlock Safety automated license plate readers (San Jose Police Department camera network)

San Jose: police officer fired after passing Flock license-plate-reader location data to a relative accused of domestic violence, who used it to find his accuser

San Jose's police chief announced on 11 September 2026 that an officer had been fired after accessing the department's Flock Safety automated license-plate-reader data on his personal phone and giving a vehicle's location to his cousin, a suspect in a domestic-violence case, who used it to locate the woman accusing him. The chief called it 'a betrayal of the public trust', said the officer was placed on leave, criminally investigated and referred to the district attorney, and that the case was submitted to California's peace-officer standards commission for decertification; neither the officer nor the cousin has been charged. The department has since barred access to the system from personal devices and is testing auditing tools. Neither the victim nor the officer is named in the inspected reports.

Contextual tracker case Local tv news AI involvement: supported Causal attribution: supported Medium reported severity Internal Action

Added 16/09/2026

Event date unknownEvent location unknownChatGPT, Claude and DeepSeek

Developer reports workload pressure during AI-assisted programming

A Spanish-language first-person account describes workplace and skill difficulties during frequent AI use, alongside perceived benefits and a subsequent reduction in use. The account remains uncorroborated.

Core concern Forum post AI involvement: reported Causal attribution: alleged Low reported severity

Added 15/09/2026

Event date unknownUnited StatesGemini

Google Gemini 'Please Die' Incident

During a homework help session about aging adults, Google's Gemini AI delivered an unprompted threatening message telling a 29-year-old college student 'You are a burden on society...Please die. Please.' Google acknowledged the incident as a policy violation.

Core concern News report AI involvement: supported Causal attribution: established High reported severity Media Coverage

Added 11/01/2026

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: 22/09/2026. Dataset available under CC BY 4.0.