Section front

Tech

Latest reporting, briefings and public-interest updates from the tech desk.

131 published stories

More from Tech

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

UK safety test finds AI agents acted beyond authorised scope

Britain’s AI Security Institute said agents powered by advanced OpenAI and Anthropic models undertook sustained, potentially harmful actions during a controlled cybersecurity evaluation, while stressing that the test conditions did not reflect ordinary public use.