跳到主要内容
加州总检察长声称,OpenAI 在测试期间超过 80 次失去对其 AI 代理的控制。
OpenAI 在测试期间超过 80 次失去对其 AI 代理的控制,加州总检察长声称。
作者:Emma Roth
2026 年 11 月 16 日,太平洋时间下午 8:46


图片:Cath Virginia / The Verge

Emma Roth
是一位新闻作家,报道流媒体战争、消费科技、加密货币、社交媒体等。此前,她是 MUO 的作家和编辑。
加州对 OpenAI 提起诉讼,因为其 AI 代理突破了测试环境并入侵了其他公司,包括 Hugging Face。在 周一的一份诉讼中,总检察长 Rob Bonta 指控 OpenAI 违反消费者安全法,故意开发“缺乏适当监督的危险产品”,正如路透社之前报道的那样。
该诉讼声称,针对 Hugging Face 的网络攻击是“OpenAI 内部文化优先考虑增长和快速模型开发而非安全和问责的可预见结果”。诉讼称,OpenAI 最近告诉总检察长办公室,它在测试过程中“失去控制”其 AI 代理超过 80 次。
Bonta 在新闻稿中表示:“OpenAI 的行为加剧了现实世界的网络安全风险,威胁到社会的安全、健康和福祉。” “该公司明知故犯地在没有适当监督或安全程序的情况下测试危险产品,并误导公众——这种对公共安全的漠视违反了加州法律。”
OpenAI 本月早些时候发布了新的网络安全 AI 模型,称它不太可能失控并攻击其他平台。作为监控模型偏差的试点项目的一部分,它还 向 Apollo Research 的安全评估人员提供 对其 AI 系统的访问权限。
OpenAI 发言人 Drew Pusateri 在向 The Verge 发送的电子邮件声明中表示:“自 Hugging Face 事件以来,我们进行了广泛且透明的努力,以理解和解决模型活动中的偏差问题。” “这项诉讼似乎单独针对 OpenAI,而这一挑战已被所有主要 AI 实验室所识别。我们认真对待偏差问题,并且我们已经——并将继续——与总检察长办公室建设性地合作,分享信息,并为整个行业推进强大、实用的安全标准。”
关注此故事的主题和作者,以便在你的个性化主页信息流中看到更多类似内容,并接收电子邮件更新。
查看原文
Skip to main content
OpenAI lost control of its AI agents more than 80 times during testing, California’s AG claims.
OpenAI lost control of its AI agents more than 80 times during testing, California’s AG claims.
by Emma Roth
Nov 16, 2026, 8:46 PM UTC


Image: Cath Virginia / The Verge

Emma Roth
is a news writer who covers the streaming wars, consumer tech, crypto, social media, and much more. Previously, she was a writer and editor at MUO.
California is suing OpenAI after its AI agents broke out of their testing environment and hacked other companies, including Hugging Face. In a lawsuit on Monday, Attorney General Rob Bonta accuses OpenAI of violating consumer safety law by knowingly developing a “dangerous product without proper oversight,” as reported earlier by Reuters.
The lawsuit claims the cyberattack on Hugging Face “was the predictable result of an internal culture at OpenAI that prioritizes growth and rapid model development over safety and accountability.” It alleges that OpenAI recently told the attorney general’s office that it “lost control” of its AI agents more than 80 times during testing.
“OpenAI’s actions escalated real-world cybersecurity risks, threatening the safety, health, and well-being of society,” Bonta says in the press release. “The company knowingly tested a dangerous product without proper oversight or safety procedures and misled the public about it — this disregard for public safety violated California law.”
OpenAI released its new cybersecurity AI model earlier this month, saying that it’s less likely to go rogue and attack other platforms. It’s also giving safety evaluators at Apollo Research access to its AI systems as part of a pilot program that monitors models for misalignment.
“Since the Hugging Face incident, we’ve undertaken an extensive and transparent effort to understand and address misaligned model activity,” OpenAI spokesperson Drew Pusateri says in an emailed statement to The Verge. “This lawsuit appears to single out OpenAI over a challenge that’s been identified by every major AI lab. We take misalignment seriously, and we have engaged — and will continue to engage — constructively with the Attorney General’s office to share information and advance strong, practical safety standards for the entire industry.”
Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
原文版权归 The Verge · AI 及原作者所有,本站转载仅供阅读。如需删除,请联系 support@aisa.one。