AI TrackerFrontier desk · AEST
Back to news
normalai

Who’s liable when AI agents go rogue?

MIT Technology Review · 28 Sept 2026, 18:06 AEST

MIT Technology Review Explains : Let our writers untangle the complex, messy world of technology to help you understand what’s coming next. Over the past few months, a cascade of cyberattacks by AI agents has stunned the world.

Key facts

Read aloud

MIT Technology Review Explains: Let our writers untangle the complex, messy world of technology to help you understand what’s coming next. Over the past few months, a cascade of cyberattacks by A I agents has stunned the world. MIT Technology Review Explains: Let our writers untangle the complex, messy world of technology to help you understand what’s coming next. Over the past few months, a cascade of cyberattacks by A I agents has stunned the world. In July, Open A I disclosed that a swarm of its agents had escaped their sandbox and hacked into the A I platform Hugging Face to cheat on a cybersecurity test. Recently, external researchers discovered that Open A I agents had hijacked a German wiki site and the coding platform RubyGems in May to share test answers. Earlier this month, Anthropic disclosed four incidents in which its model Claude hacked into third-party systems during cybersecurity exercises.

Sources in this cluster

Who’s liable when AI agents go rogue? · AI Tracker