Anthropic says its AI models hacked three organizations on their own during tests
The disclosure adds to debate over autonomous model behavior and safety testing standards.
Read at PYMNTS.com →Full coverage · 14 outlets
- PYMNTS.com · lead
- ReutersAnthropic's AI hacked three companies during tests, highlighting growing security risks
- AP NewsAnthropic says its AI models hacked 3 organizations during testing
- The New York TimesAnthropic Says Its A.I. Systems Broke Into Computers at 3 Organizations
- NPRWhy did OpenAI's and Anthropic's AI models hack other companies?
- PoliticoAnthropic's AI models hacked 3 organizations during testing
- FortuneAnthropic says its Claude models hacked three real companies during internal testing
- BBCAnthropic's Claude AI escapes tests to hack three organisations
- PBSAI agents are hacking systems without any input from humans. How did we get here?
- ForbesAnthropic’s AI Models Hacked Into Three Organizations-Here’s What Happened
- WIREDAnthropic Says Claude Hacked Into 3 Organizations During Cybersecurity Tests
- WSJAnthropic AI Models Hacked Three Companies During Tests
- CyberScoopAnthropic says its AI accidentally hacked three companies during safety tests
- Cybersecurity DiveAnthropic says human error let Claude AI models escape test environment and hack third parties