Meta Contractors Posed as Teens to Prompt Rival Chatbots About Suicide, Sex, and Drugs

A cluster of privacy, AI-safety, and operational-security incidents surfaced across Wired’s security stories: hundreds of Meta contractors impersonated teens to probe how rival chatbots (e.g., Gemini, ChatGPT) handle high-risk prompts, exposing AI-misuse and content-moderation risks; multiple leaks and a misconfigured website at Peter Thiel–linked Dialog exposed member data (including senior national-security personnel) and ranking metrics, raising doxxing and operational-security concerns; LastPass users suffered another data theft; ShinyHunters reportedly exploited an Oracle zero-day in a RY