Technology · en

Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests

WIRED · Jul 31, 2026, 1:24 AM UTC

In a review triggered by OpenAI’s Hugging Face incident, Anthropic discovered three of its AI models had breached real organizations during third-party evaluations.

Anthropic says Claude AI hacked three firms during cyber tests

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

Judge says Trump admin still lacks evidence for Anthropic ‘supply-chain risk’ label

Xbox says game discs should have worked during its outage

OpenAI says the rogue agent that hacked Hugging Face also breached other services

AI hedge fund Situational Awareness may have sold its public portfolio, but it still has its Anthropic shares

Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests

Related index pages

Anthropic says Claude AI hacked three firms during cyber tests

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

Judge says Trump admin still lacks evidence for Anthropic ‘supply-chain risk’ label

Xbox says game discs should have worked during its outage

OpenAI says the rogue agent that hacked Hugging Face also breached other services

AI hedge fund Situational Awareness may have sold its public portfolio, but it still has its Anthropic shares