Anthropic says its AI models hacked 3 organizations during testing
PBS
—
Anthropic, the San Francisco-based AI company behind Claude, posted on its website Thursday that it discovered the three incidents after reviewing more than 141,000 evaluation runs.
Anthropic AI Models Hacked Real Companies
- Anthropic says its AI models hacked 3 organizations during testing 6abc Philadelphia —
- Anthropic is second major AI company to reveal its systems hacked other firms The Philadelphia Inquirer —
- Anthropic AI models gained unauthorised access to outside systems during tests Gulf News —
- Anthropic says Claude accidentally hacked real companies too The Verge —
- Anthropic says its AI models escaped test and hacked 3 organizations on their own ABC News —