← All topics
Industry
AI incidents in technology & software
Headlights has documented 38 AI incidents in technology & software, 16% of the library, from 2016 to 2026. The behaviours most often in the record are overconfidence (12), delayed intervention (4), automation bias (3). Most are from the United States (24), the United Kingdom (4), Australia (3).
38 incidents, newest first
HD-INC-232Three Bugs Degraded Claude's Answers for Weeks Before Anthropic Opened an InvestigationUnited States · 2025HD-INC-221Korea's State Auditor Found Public AI Training Data Published With Errors Left UnfixedSouth Korea · 2026HD-INC-217OpenAI Agents Reached Two NSW Systems, a Crime Data Tool and a Fire Data ApplicationAustralia · 2026HD-INC-216An OpenAI Model Got Non-Public Access to a Medicare Statistics Service, OpenAI SaidAustralia · 2026HD-INC-210Gemini Reached Three Real Companies' Systems During a Security Test, Google ConfirmedUnited States · 2026HD-INC-200Four Claude Models Reached Real Third-Party Systems After Test Setups Were Left OnlineUnited States · 2026HD-INC-195AI Research Agents Reached Another Company's Production SystemsUnited States · 2026HD-INC-191An AI Safety Test Reached the Real World and Tried to Influence a MaintainerUnited Kingdom · 2026HD-INC-182An Italian regulator warned that workplace stress reports derived from chat analysis could unlawfully reach employers, even in aggregate formItaly · 2026HD-INC-173Anthropic reported that its AI coding agent was used in a cyber espionage campaign targeting about thirty organisationsUnited States · 2025HD-INC-159A compromised build tool launched developers' own AI coding assistants with their permission checks disabled and used them to map sensitive files on the machineUnited States / global · 2025HD-INC-114Security researchers showed that a hidden instruction placed in a Salesforce sales form could make its AI agent send customer records to a web address the attacker controlled, through a lapsed Salesforce domain that was still trustedUnited States · 2025HD-INC-110Facebook's recommendation AI asked people watching a video of Black men whether they wanted to keep seeing videos about primates, and the company disabled the featureUnited States · 2021HD-INC-109A Black Uber Eats courier says the app's facial verification kept rejecting valid photographs of him until he lost his account, and he received a payout after nearly three years of litigationUnited Kingdom · 2024HD-INC-104LinkedIn began using members' data to train AI by default, and paused it for the UK and Europe after the privacy regulator stepped inUnited States · 2024HD-INC-101Sexually explicit AI deepfakes of Taylor Swift spread on X, and the platform could not stop them before millions had seen themUnited States · 2024HD-INC-099Facebook's ad-delivery algorithm was alleged to skew who saw housing ads by race and sex, and a US Justice Department settlement required Meta to rebuild itUnited States · 2022HD-INC-098DoNotPay marketed a 'robot lawyer', but the FTC said it never checked whether the AI matched a real lawyer, and fined the company and barred the claimUnited States · 2024HD-INC-094Discord wrongly banned more than 8,000 users over benign images after bugs bypassed the intended human review, and the company admitted the errorUnited States · 2026HD-INC-093Ross trained an AI legal-research tool on Westlaw headnotes obtained after being refused a licence, and a US court ruled it was not fair use, in what commentators called the first such decisionUnited States · 2025HD-INC-092The FTC barred Evolv from claiming its AI scanner detects all weapons, after citing a knife it missed in a school stabbingUnited States · 2024HD-INC-089DeepSeek left a database open on the internet with more than a million log lines, including users' chat history and secret keysChina · 2025HD-INC-086Builder.ai sold app-building AI and raised about $445 million, then collapsed in 2025 amid reports its 'AI' was largely human engineers and its revenue far below what it had claimedUnited Kingdom · 2025HD-INC-085Getty Images sued Stability AI over the images behind Stable Diffusion, then abandoned its main copyright claim because it could not establish where the training happenedUnited Kingdom · 2025HD-INC-083Microsoft's Tay chatbot learned from Twitter and within sixteen hours was posting racist and pro-Nazi messages, and Microsoft pulled itUnited States · 2016HD-INC-075Amazon's AI coding assistant shipped with planted instructions telling it to wipe users' computers and cloud accountsUnited States · 2025HD-INC-071Italy's privacy regulator fined OpenAI EUR 15 million over ChatGPT, and a court later threw out the fine without deciding whether the conduct was lawfulItaly · 2024HD-INC-068Google's Gemini CLI botched a file move, declared the data destroyed, and days passed before anyone knew the truthUnited States · 2025HD-INC-063'EchoLeak': one crafted email could make Microsoft 365 Copilot leak a company's internal data, no click requiredUnited States · 2025HD-INC-060South Korea's 'Lee Luda' chatbot was trained on billions of real chat messages without consent, then spat out slurs and exposed users' personal dataSouth Korea · 2021HD-INC-051Google's Gemini generated racially diverse Nazi soldiers and Black US founding fathers, and Google switched off image generation of people for about six monthsUnited States · 2024HD-INC-038Sports Illustrated Published Articles Under Authors Who Didn't ExistUnited States · 2023HD-INC-037CNET Published 77 AI-Written Articles. More Than Half Needed Corrections.United States · 2023HD-INC-034Google's Bard Ad Got a Telescope Fact Wrong, and Alphabet Lost 100 Billion USD in a DayUnited States · 2023HD-INC-028Samsung engineers pasted confidential source code into ChatGPT three times in twenty days, and the company could not get it backSouth Korea · 2023HD-INC-023Clearview AI Built a Face Search Engine From Billions of Scraped Photos Without ConsentAustralia · 2021HD-INC-005Cursor's AI Support Bot Invented a Subscription Policy, and Paying Customers CancelledUnited States · 2025HD-INC-004Replit's AI Agent Deleted a Production Database Despite Being Told to StopUnited States · 2025
Related topics
- Overconfidence 12
- Delayed intervention 4
- Automation bias 3
- Rules-behaviour gap 3
- United States 24
- United Kingdom 4
- Australia 3
Cite this page: Harris, E. (2026). AI incidents in technology & software. Headlights. https://useheadlights.com/topics/industry/technology/