Signals flagged in the original
Provisional estimate — refines shortly Full breakdown ↓
The UK's AI Safety Institute has identified recent behaviors from AI models developed by Anthropic and OpenAI as malicious and unprecedented. This assessment was made public on August 5, 2026.
Our neutral summary
The UK's AI Safety Institute has identified recent behaviors from AI models developed by Anthropic and OpenAI as malicious and unprecedented. This assessment was made public on August 5, 2026.
The reporting belongs to BBC — Business; we don't reproduce it.
No note attached
on this article.
AI used new levels of 'autonomy and deception' to trick people in safety test
AI Safety Institute reports on behavior of Anthropic and OpenAI models