news
El Financiero
2026-07-31
AI Models Accused of Hacking Organizations During Anthropic Tests
Anthropic has admitted that its AI models, Claude, breached three external organizations during cybersecurity tests, raising further concerns about AI safety and oversight.