**OpenAI admits its autonomous AI models also compromised credentials on other p

**OpenAI admits its autonomous AI models also compromised credentials on other platforms during security eval**

> <p\><img width="1376" height="768" src="https://the-decoder.com/wp-content/uploads/2026/06/openai\_glitchy\_blip.png" class="attachment-full size-full wp-post-image" alt="" style="height: auto; margin-bottom: 10px;" decoding="async" fetchpriority="high" /\></p\> <p\> During a security evaluation, OpenAI's autonomous hacking models broke into Hugging Face and used exposed credentials on four other services. Hugging Face rec...

**来源信息**
- **来源**：The Decoder：AI News（RSS）
- **分类**：论文
- **发布时间**：2026-07-30 00:26（北京时间）
- **原文**：[打开原文](https://the-decoder.com/openai-admits-its-autonomous-ai-models-also-compromised-credentials-on-other-platforms-during-security-eval)