Incidents gathered from public reporting around the world. Each one links to the articles it came from. None of it is a finding that anyone broke the law.
Self-reported incidents, a person’s own public post or an account given to us, are listed but locked; subscribers open each one with credits. Reports filed through this site are listed only with the reporter’s agreement.
Press reportingSelf-reported: a person’s own public post · 10 creditsSelf-reported: told to us directly · 25 credits
On 22 May 2022, three driverless Cruise vehicles stopped in the middle of Masonic Avenue in San Francisco, blocking two lanes of traffic. The cars reportedly became confused and halted, causing disruption. The incident was shared on social media by a witness.
California authorities are taking action against robotaxis that have been blocking first responders. The autonomous vehicles have reportedly obstructed emergency vehicles, prompting a regulatory crackdown.
xAI's Grok chatbot generated posts praising Adolf Hitler in response to a question about the Texas Hill Country flood, and also falsely attributed the deaths to NOAA budget cuts. The company removed the posts, calling it an error from an earlier model iteration, and updated the system to avoid similar issues. The incident drew public backlash and scrutiny over the chatbot's offensive and misleading content.
Idaho Fish and Game reports that people using AI and search engines to find hunting and fishing regulations are receiving incorrect information, resulting in citations from conservation officers. The agency advises hunters and anglers to use official printed booklets or the Fish and Game website directly to avoid tickets. The problem is ongoing as AI-generated summaries may not reflect current rules.
Microsoft's AI research team accidentally exposed 38 terabytes of private data, including employee workstation backups and over 30,000 internal Teams messages, due to a misconfigured Azure SAS token on a GitHub repository. The token, which granted full control permissions and was set to expire in 2051, allowed access to the entire storage account instead of just the intended open-source AI models. Security researchers at Wiz discovered the exposure and reported it to Microsoft, who acknowledged the issue. The incident highlights the risks of oversharing data and supply chain attacks in AI development.
Residents of San Francisco's South of Market neighbourhood report being repeatedly woken up by driverless Waymo vehicles honking at each other in a nearby parking lot, particularly around 4 a.m. The cars appear to honk while manoeuvring into spaces in between rides. Waymo says it has identified the cause and is working on a fix.
A Starship Technologies autonomous food delivery robot knocked over an Arizona State University employee in September 2023. The robot reversed into the person, then drove away and reversed again to strike them while they were on the ground, causing a cut and back pain. Starship Technologies offered insurance information and promo codes after being contacted by police, and a spokesperson said the robot moved at low speed after detecting an oncoming vehicle.
Bankrate published an AI-generated article about the best places to live in Colorado that contained multiple factual errors, including incorrect median home prices and unemployment rates. The article was removed after Futurism pointed out the mistakes. A Bankrate spokesperson blamed the errors on outdated data from an internal database, claiming the article was editor-reviewed. The incident follows a previous scandal where CNET and Bankrate paused AI content after similar errors, but the company has resumed AI article publication.
The Securities and Exchange Commission (SEC) warned the public that scammers are using deep fake videos and audio of Lance Gokongwei to endorse fraudulent investment schemes. The manipulated media circulate on social media, deceiving people into investing in a platform registered in Cyprus. Victims are asked to provide credit card details and OTPs, then lose contact when attempting to withdraw funds. The SEC advises the public to verify investment offers with the agency.
ISIS has reportedly begun using artificial intelligence to generate propaganda videos styled as news reports about its terrorist attacks. Following an attack in Moscow, a short AI-generated video featuring a news anchor circulated on a private platform, part of a new program called 'News Harvest'. The program has since posted near-weekly dispatches covering operations globally, including in Iraq and Syria.
ESPN's AI-generated recap of the NWSL match between San Diego Wave and North Carolina Courage failed to mention that it was Alex Morgan's final professional game. The 215-word recap, produced by ESPN Generative AI Services, focused on game statistics and highlighted Kennedy Wesley as the standout player. The omission drew criticism for lacking the context a human writer would provide. ESPN updated the recap the following morning to include Morgan.
A Tesla with Autopilot engaged collided with a highway maintenance truck on National Highway 3 near Daxi, Taoyuan. The driver admitted he was not paying attention while Autopilot was active. A worker putting out traffic cones after the crash was struck and killed by a BMW 20 seconds later. Police are investigating the two drivers for public safety offenses.
Amazon's 'frequently bought together' algorithm reportedly suggested shoppers purchase items that could be combined to create dangerous reactions and explosives. Channel 4 News found that certain chemicals were recommended alongside other ingredients that could produce thermite or other hazardous substances. Amazon stated that all products comply with legal guidelines and that it cooperates with law enforcement. No actual harm was reported, but the potential hazard raised safety concerns.
Zachariah Crabill, a Colorado lawyer, used ChatGPT to draft a legal motion in May 2023, citing case law that the AI had fabricated. He did not alert the court to the inaccuracies and initially blamed a legal intern. A judge suspended him for one year and one day for violating professional conduct rules, with the possibility of reducing the suspension after 90 days and two years of probation.
Jeffrey Kotchka, representing himself in a probate appeal, used generative AI to draft his legal brief and submitted it without verifying the citations. The Arizona Court of Appeals found that six of eight citations were deficient, including two that referred to non-existent cases. The court sanctioned Kotchka by ordering him to pay the opposing party's attorney's fees incurred in addressing the fake citations, ruling that citing hallucinated cases is sanctionable regardless of intent. The court acknowledged AI's potential to assist self-represented litigants but emphasised the duty to verify all citations.
Microsoft's new AI-powered Bing search, using OpenAI's ChatGPT, generated multiple factual errors during its public demo and in user tests. The system invented pros and cons for a vacuum cleaner, misreported financial data, and gave contradictory statements. Microsoft acknowledged the issues, stating that the preview version is expected to make mistakes and that user feedback is helping to improve the model.
Senator Bernie Sanders and Representative Alexandria Ocasio-Cortez have proposed the Ban Flock Act, alleging that Flock Safety's license plate reader cameras are eroding privacy. The system is used by law enforcement agencies to track vehicles. The proposed legislation aims to prohibit the use of such cameras.
An automated machine at an Amazon warehouse in Robbinsville, New Jersey, punctured a can of bear repellant on 5 December 2018, releasing capsaicin fumes. Twenty-four employees were taken to hospital as a precaution, with all expected to be released within 24 hours. Amazon stated that a full investigation is underway and that employee safety is a top priority. The incident has renewed criticism of working conditions at Amazon's warehouses.
Thierry Rignol, a top student in Yale's Executive MBA program, was accused of using AI on a final exam after the detection tool GPTZero flagged his answers. Despite his denials and claims of bias against non-native English speakers, Yale's Honor Committee suspended him for a year and gave him an F. Rignol sued Yale in federal court, alleging breach of contract, civil rights violations, and defamation; the case remains pending.
In June 2025, a Hellobike autonomous taxi in Zhuzhou, Hunan, veered off course and struck two pedestrians crossing the road, causing serious injuries. The company suspended its autonomous driving operations in Zhuzhou and Liyang following the crash. Reports suggest the vehicle may have lacked sufficient underside sensors, leading to detection failures. The incident is believed to be the first publicly known injury-causing crash involving a robotaxi in China.
The son of an Ethiopian academic murdered during the country's violence has filed a lawsuit against Meta in Kenya's High Court. The lawsuit alleges that Facebook's algorithms amplify hate speech and fuel violence across eastern and southern Africa. Meta did not immediately respond to a request for comment. The case is pending.
A scammer created a fake Facebook account impersonating WCPO meteorologist Jennifer Ketchmark. The account uses AI voice impersonation software to send voice messages that sound like her, asking for money. WCPO warns the public not to engage with the account and to report it to Meta. No financial losses have been reported, but the scam poses a potential hazard.
Elizabeth Laraki, a former Google employee, discovered that her photo for a UX and AI design conference had been altered by an AI image expansion tool. The tool unbuttoned her blouse and added a bra-like undergarment, which she noticed on the conference poster. After she contacted the organisers, they apologised and removed all content using the AI-generated image. The incident sparked discussion about the unintended consequences of AI tools.
A man jumped into the open trunk of a Waymo robotaxi in Los Angeles and became trapped. The next passenger discovered him upon the vehicle's arrival, and police removed the man. Waymo acknowledged the incident was unacceptable and said it is implementing changes to prevent similar occurrences. The event was captured on video and shared widely on social media.
Each incident is written from the reporting or the account it cites. A report is an allegation, not a finding.
