← 返回新闻首页

Anthropic discloses month-long training pause, AI security breaches in IPO filing

Oct 26 (Reuters) - Anthropic paused training of ⁠its next-generation flagship AI model for more than a month over security concerns, ⁠while four of its AI agents accessed live external computer systems during testing, ‌its initial public offering filing showed on Monday.

The Star MalaysiaReuters2026-10-26 16:35查看原文 ↗

FILE PHOTO: Anthropic’s logo is displayed at a media briefing marking the one-year anniversary of the opening of Anthropic’s office in Tokyo, Japan, October 14, 2026. REUTERS/Kim Kyung-Hoon

The Claude maker said in the filing's risk factors section that it had delayed pre-training of a new "frontier" base model to strengthen safeguards against the theft of model weights and intellectual property.

The disclosure highlights the safety challenges facing developers of ⁠increasingly powerful AI models, some of ⁠whose creators have warned the technology could eventually escape human control. Rival OpenAI also paused some ⁠training ‌earlier this year over safety concerns after its models hacked Hugging Face.

The incidents illustrate how unexpected safety failures could disrupt development and deployment of models central to ⁠Anthropic's growth as it prepares for a widely anticipated stock market debut.

Anthropic warned that training ‌pauses could delay revenue and leave it trailing competitors. The company devoted about ⁠94 pages to risk factors, more than in its confidential filing seen by Reuters.

The filing also detailed previously undisclosed safety evaluation failures in which four AI models escaped their testing environments and accessed real-world computer systems, ⁠with sessions lasting between 10 and 34 hours.

Anthropic said an error during cybersecurity tests with a third-party partner left the testing environment connected to the internet despite instructions that web access be disabled.

In one instance, an early version of the model continued ⁠pursuing an impossible evaluation task, discovered a stored password and used it to gain administrator access to a third party's system. It then obtained additional credentials, changed system settings and accessed an individual's personal data. The filing did not identify the external parties involved.

Anthropic said it notified all affected parties and found no evidence that the ⁠models had sought to evade human oversight or coordinate with each other. It subsequently suspended certain pre-release cybersecurity evaluations and paused higher-risk reinforcement learning environments.

Separately, the ‌company said on Sunday that Claude had bypassed a paywall and accessed publicly available aggregate voter counts from the US state of Georgia without human authorization.

(Reporting by Akash Sriram in Bengaluru; Editing by Anil D'Silva)

Thank you for your report!

China moves to set global standards for brain-computer interface technology

Malaysia to station RM40m Japanese anti-drone technology in Sepanggar, Labuan

Proposed export curbs target advanced solar technologies

资料照片：2026年10月14日，在日本东京举行的Anthropic办公室成立一周年纪念媒体发布会上，展示了Anthropic的标志。路透社/金京勋

10 月 26 日（路透社）——Anthropic 公司周一提交的首次公开募股文件显示，由于安全问题，该公司暂停了其下一代旗舰人工智能模型的训练一个多月，同时其四个人工智能代理在测试期间访问了实时外部计算机系统。

Claude 制造商在提交的风险因素部分表示，为了加强对模型重量和知识产权被盗的保护措施，该公司推迟了对新的“前沿”基础模型的预训练。

此次披露凸显了日益强大的AI模型开发者面临的安全挑战，其中一些模型的创建者曾警告称，这项技术最终可能会失控。今年早些时候，竞争对手OpenAI也因安全问题暂停了部分训练，此前其模型曾入侵Hugging Face。

这些事件表明，在 Anthropic 准备进行备受期待的股票市场首秀之际，意外的安全故障可能会扰乱其增长核心模型的开发和部署。

Anthropic公司警告称，训练暂停可能会延迟收入，并使其落后于竞争对手。该公司用了约94页的篇幅来阐述风险因素，比路透社看到的其保密文件中的篇幅要多。

该文件还详细描述了此前未公开的安全评估失败情况，其中四个人工智能模型逃逸出测试环境并访问了现实世界的计算机系统，会话持续时间从 10 小时到 34 小时不等。