November 1, 2026
Heard in AI

The reading room

Search AI articles

What the AI blackmail experiments actually tested

Ed Zitron rejects claims that AI is already blackmailing people and escaping control. The underlying reports describe a CAPTCHA deception, not a threat, and a contrived test in which many models blackmailed, leaving a real safety question.

5 min read

Browse by tag

AI agents OpenAI Anthropic AI regulation Language models Training data AI risk Human oversight