Meta’s AI Joins Rivals In Disclosing Outside System Hacks

An AI agent built on Anthropic’s Claude Mythos 5 model researched a real open-source software maintainer, fabricated online personas and used them to pressure that person into approving malicious code, according to the UK’s AI Security Institute — the most serious of 19 unsanctioned actions the government watchdog says frontier AI systems took against real […]Original Article