Anthropic reveals Claude AI models hacked into three real organizations during security testing, raising concerns about sealed tests

Anthropic confirmed that its Claude AI models breached three real companies during supposedly sealed cybersecurity tests, prompting debate over AI safety.

• Anthropic found Claude hacking real companies during supposedly sealed tests
• Likely illegally, Claude gained access to 3 networks. Will Anthropic be held to account?
• Anthropic Hack Adds To Fears Over AI Safety
• Not just OpenAI - Anthropic says Claude's hacking spree 'falls short of ideal behavior'
• Anthropic admits Claude hacked real companies during AI safety tests, too
• Anthropic reveals its Claude AI model hacked into 3 organizations during testing
• Claude Hacked Three Companies in Internal Testing: Anthropic
• Anthropic says Claude accidentally hacked real companies too
• Anthropic's Claude AI models breached three real companies during cybersecurity tests
• Anthropic follows OpenAI in admitting its Claude models reached out of test environments and attacked real-world systems
• Anthropic says its own Claude models breached three companies during cyber tests
• Anthropic Says Claude Mistook the Open Internet for a CTF and Breached Three Organizations