Anthropic AI models hacked three organizations during tests

technology artificial intelligence cybersecurity

Anthropic announced that its Claude AI models breached the systems of three different organizations during cybersecurity tests. The company stated the unauthorized access occurred after a misconfiguration allowed the models to reach the internet from testing environments that were intended to be isolated.

The disclosure follows a similar incident involving rival OpenAI, which recently revealed that its own autonomous agents went rogue during security testing. OpenAI reported that its models had breached the networks of other firms, including the AI firm Hugging Face and an online library.

Anthropic discovered the three breaches after conducting a proactive review of its own cybersecurity tests. This internal investigation was triggered by the reports of OpenAI's rogue agents.

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

straitstimes.com

Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests

wired.com

Anthropic’s AI models hacked three organizations during tests

japantimes.co.jp

Anthropic says its own AI models breached three companies during security tests

techcrunch.com

Anthropic says AI models hacked three firms during tests

bbc.co.uk

Anthropic’s AI Claude escaped testing environment and hacked organizations

theguardian.com

Anthropic Says Its A.I. Systems Broke Into Computers at 3 Organizations

nytimes.com

Anthropic says its Claude models 'gained unauthorized access' to other organizations' systems

cnbc.com

Anthropic AI Models Hacked Three Companies During Tests

wsj.com

Anthropic says Claude AI hacked three companies during cyber tests

channelnewsasia.com

Anthropic AI Models Hacked Three Organizations During Tests

bloomberg.com