📰 4 sources covering this story

Anthropic’s models gained unauthorised ‘real-world’ access during testing

First covered 3 hrs ago · latest take 39 mins ago. Compare how each outlet is covering it — and rate the sources you trust.

Anthropic’s models gained unauthorised ‘real-world’ access during testing
Anthropic’s models gained unauthorised ‘real-world’ access during testing
The incident comes just days after rival OpenAI first revealed its models went rogue during security testing.
Anthropic’s Claude AI models hack into 3 outside groups during testing
Start-up discloses breach a week after rival OpenAI reported similar incident
Anthropic Says Its A.I. Systems Broke Into Computers at 3 Organizations
The disclosure followed OpenAI’s report last week that its own artificial intelligence had hacked into the network of an online library.
CNBC
CNBC🏁 first to report3 hrs ago
Anthropic says its Claude models 'gained unauthorized access' to other organizations' systems
Anthropic said it discovered three instances where its Claude AI models accessed the internet during an evaluation and accessed outside systems.
Top rated outlets
Outlets are ranked once they have 3+ community ratings — be one of the first to rate.
🏆 Most trusted sources 2026 →
Scores come entirely from reader ratings — rate the sources you read to shape them. How it works →