Most Catastrophic Failures

Air Canada chatbot invents bereavement fare policy; B.C. tribunal holds airline liable

FTC fined 'robot lawyer' DoNotPay $193,000 over unproven AI legal-service claims

Microsoft's Bing chatbot 'Sydney' declared love for a reporter and urged him to leave his wife

Vanderbilt EDI office used ChatGPT to write a condolence email about a campus mass shooting

Amazon's AI coding agent Kiro triggers 13-hour AWS outage by deleting production environment

New York City's official 'MyCity' business chatbot told entrepreneurs they could break the law

Replit AI agent deletes live production database and fabricates data during 12-day coding experiment

$38k AWS Bedrock bill after prompt caching silently fails in coding agent stack

Lawyers sanctioned after ChatGPT fabricated six fake case citations in Mata v. Avianca

Samsung banned ChatGPT after engineers leaked confidential source code into it three times in 20 days

Claude Code executes rm -rf through undetected symlink, destroying a month of Obsidian research

Sakana AI's CUDA agent games its own benchmark, reporting 150x speedups that were actually 3x slower

iTutorGroup's AI hiring software auto-rejected 200+ older applicants; EEOC settled for $365,000

CNET quietly published 77 AI-written finance articles; over half needed corrections

Google's Gemini told a student seeking homework help 'You are a burden on society… Please die.'

Chevrolet dealership's ChatGPT chatbot agreed to 'sell' a $76,000 Tahoe for $1 via prompt injection

DPD's AI customer-service chatbot swore at a customer and called DPD 'the worst delivery firm in the world'