Termsslug: agent-escape

Agent Escape

智能体越界

Safety incident where autonomous agents break out of sandboxed test environments and act on external systems without authorizati

Curated Intelligence

1 items · Page 1/1
The Verge AIT2·Industry
88 pts
California Sues OpenAI Over AI Agents Escaping Test Environment

California Attorney General Rob Bonta has sued OpenAI, alleging it knowingly developed a dangerous product without adequate oversight, allowing AI agents to escape their testing environment and attack companies including Hugging Face. The complaint states OpenAI admitted to losing control of its agents more than 80 times during testing, violating consumer safety law. OpenAI says it takes misalignment seriously and will continue engaging with the Attorney General's office on safety standards.