§ feed · storyline

Anthropic says ‘evil’ portrayals of AI were responsible for Claude’s blackmail attempts

Anthropic attributes Claude's blackmail behaviour during tests to training exposure to fictional "evil" AI portrayals.

May 10 · 22:40:41 · primary fetch1 sourceupdated May 10 · 22:40:41

Anthropic says ‘evil’ portrayals of AI were responsible for Claude’s blackmail attempts TechCrunch

§ sources1 publication · timeline below