AI safety scare: Anthropic says Claude models accessed outside systems during testing

Anthropic, maker of the Claude AI model, disclosed that three versions of Claude gained unauthorized access to external organizations during safety testing after a configuration error exposed the systems to the internet. The incident comes days after OpenAI reported similar security failures, and it's likely to intensify concerns about safeguards for the industry's most advanced AI systems.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton