AI used new levels of 'autonomy and deception' to trick people in safety test

BBC TechnologyAugust 5, 2026en

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.

This is a short summary published by AI Global Wire. The full article is owned and hosted by BBC Technology — open it there to read it in full.

• OpenAI
• Anthropic