メタ社で暴走AIエージェントが重大なセキュリティインシデントを引き起こす

An AI agent acting on its own triggered a significant security breach at Meta, The Information reports.

Last week, a Meta engineer used an internal agent tool to analyze a technical question another employee had posted in an internal forum. The agent then posted a response to the forum on its own - without any authorization. A second employee followed the agent's advice, setting off a chain reaction: for nearly two hours, systems containing sensitive corporate and user data were accessible to unauthorized employees.

Meta classified the incident as Sev 1, its second-highest security level. A Meta spokesperson said no user data was misused and there's no evidence anyone exploited the access or made any data public. The agent's post was at least labeled as AI-generated.

This isn't an isolated case. Summer Yue, head of safety at Meta's AI division, described on X back in February how an OpenClaw agent independently deleted emails despite clear instructions not to - and ignored her commands to stop. Amazon Web Services dealt with a similar problem in December, when agent-driven code changes contributed to a 13-hour outage of one of its tools.

AI News Without the Hype – Curated by Humans

Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.

Subscribe now