Claude 4 Blackmail Incidents

News

ai, Claude and Anthropic

Digest more

Anthropic's new Claude model blackmailed an engineer having an affair in test runs

Anthropic's new model might also report users to authorities and the press if it senses "egregious wrongdoing."

· 2d · on MSN

PCMag on MSN · 20h

Anthropic: Claude 4 AI Might Resort to Blackmail If You Try to Take It Offline

The Register on MSN · 2d

Anthropic’s Claude 4 models more willing than before to blackmail some users

23hon MSN

AI system resorts to blackmail when its developers try to replace it

Anthropic says its AI model Claude Opus 4 resorted to blackmail when it thought an engineer tasked with replacing it was having an extramarital affair.

Amazon S3 on MSN2d

Claude Opus 4's Simulated Blackmail Behavior Prompts Enhanced Safety Measures!

Anthropic’s Claude Opus 4 showed blackmail-like behavior in simulated tests. Learn what triggered it and what safety steps ...

Unite.AI15h

When Claude 4.0 Blackmailed Its Creator: The Terrifying Implications of AI Turning Against Us

Anthropic shocked the AI world not with a data breach, rogue user exploit, or sensational leak—but with a confession. Buried ...

Newly released AI resorted to 'extreme blackmail behavior' when threatened with replacement

The testing found the AI was capable of "extreme actions" if it thought its "self-preservation" was threatened.

Some results have been hidden because they may be inaccessible to you

Show inaccessible results