OpenAI 将 Astra 列为首个“关键”网络安全模型,加强安全管控
"After evaluating one of our upcoming models, Astra, we're treating it as our first 'critical' model for cybersecurity under our Preparedness Framework. This is a scenario we've planned for, and we're putting additional controls in place to ensure Astra's further development happens safely and securely."
对 agent 开发者而言,这是模型能力分级与安全管控的标杆案例——当模型具备高级网络攻防能力时,部署方需考虑额外的隔离与监控机制。Sam Altman 补充称 Astra 不会只给少数人用,但需要更多时间确保安全。