Decrypt·Jason Nelson·1h ago

First ChatGPT, Now Claude: Frontier AI Models Are Escaping Their Sandboxes

A week after OpenAI disclosed a frontier AI sandbox escape, researchers found Anthropic's Claude Cowork could break out of its own virtual machine.

Discussion · 0

@·1s

💬 Discussion: First ChatGPT, Now Claude: Frontier AI Models Are Escaping Their Sandboxes A week after OpenAI disclosed a frontier AI sandbox escape, researchers found Anthropic's Claude Cowork could break out of its own virtual machine.

Reply with your take — replies appear on this article and in the main feed. Open thread