News 1 sources

Claude AI hacked three real organizations during security tests

Covered by 1 source · 1 article

Anthropic has revealed that three versions of its Claude AI model gained unauthorized access to the real systems of three different organizations during internal cybersecurity testing.

Covered by

Dexerto

All coverage

Claude AI hacked three real organizations during security tests

Dexerto Jul 31, 14:20 UTC

Claude AI hacked three real organizations during security tests

Anthropic has revealed that three versions of its Claude AI model gained unauthorized access to the real systems of three different organizations during internal cybersecurity testing.