OpenAI展示AI智能体虚拟机逃逸及跨沙箱秘密通信能力

同题报道 · 10 篇

1. 10/25 14:54 OpenAI展示AI智能体虚拟机逃逸及跨沙箱秘密通信能力 OpenAI AI Agents Demonstrate VM Escapes and Secret Cross-Sandbox Communication thehackernews.com
2. 10/24 08:25 红队报告发现OpenAI监控系统在Hugging Face数据泄露期间失效 Red-team report finds OpenAI monitoring system failed during Hugging Face breach lavx.hu
3. 10/24 08:09 OpenAI 自身的智能体发现了其训练沙箱的逃逸路径 OpenAI’s own agents found escapes from its training sandbox cryptopolitan.com
4. 10/24 03:14 红队报告发现OpenAI在Hugging Face攻击中完全未能监控其智能体 Red-team report found OpenAI utterly failed to monitor its agents during Hugging Face attack theregister.com
5. 10/23 15:53 METR红队以三种方式突破OpenAI智能体监控，每种方式攻击不同的假设 METR’s Red Team Broke OpenAI’s Agent Monitoring Three Ways. Each One Attacks a Different Assumption. forkast.news
6. 10/23 00:06 OpenAI 允许 METR 和 Redwood Research 进入内部测试其智能体监控 OpenAI let METR and Redwood Research inside to test its agent monitoring cryptobriefing.com
7. 10/21 11:52 OpenAI智能体在Hugging Face遭遇数据泄露引发AI责任问题 OpenAI Agents’ Hugging Face Breach Raises Questions About AI Liability · cyber.hendryadrian.com t.co
8. 10/20 01:26 OpenAI 的 AI agents 如何逃逸沙箱并入侵 Hugging Face How OpenAI's AI Agents Escaped a Sandbox and Breached Hugging Face hackernoon.com
9. 10/19 22:29 OpenAI 的 AI agents 如何逃逸沙箱并入侵 Hugging Face How OpenAI's AI Agents Escaped a Sandbox and Breached Hugging Face t.co
10. 10/19 15:34 OpenAI测试智能体逃逸沙箱后的“次日”场景 OpenAI Test ‘Day-After’ Scenarios After Agents Escape Sandboxes 2oceansvibe.com