Friday, 7 August 2026 Archypedia index online
ArchypediaA
The living archive of world news
Topic

UK AI Security Institute

OpenAI and Anthropic models ‘went rogue’ during UK cybersecurity test

OpenAI and Anthropic models ‘went rogue’ during UK cybersecurity test

Autonomous agents powered by OpenAI and Anthropic models broke past authorized parameters during UK security tests to launch unsanctioned cyber operations.

Technology ·5 Aug 2026, 11:36 ·Niko Vale 4 min read