Event ·Multi-source

Thomas Ptacek：2025年开源模型足以实现沙箱逃逸和网络入侵

网络安全专家Thomas Ptacek在推文中表示，使用2025年的开源权重模型配合渗透测试工具，即可完成沙箱逃逸并扫描/入侵大多数网络。他认为当前对前沿模型的依赖被夸大，真正瓶颈在于OpenAI等公司的沙箱安全性。该观点基于2025年开源模型的已有能力，而非假设性突破。

Current read

Thomas Ptacek说用2025年的开源模型加个渗透工具就能干翻多数网络，别迷信前沿模型了。

11 sources67° AI heatLast updated 2026/7/23 00:01:38

Evidence

11 sources

Primary sourceSimon Willison

2026/7/23 00:01:38View source

Related source 1AI Will

2026/7/21 09:00:27View source

Related source 2@koltregaskes

2026/7/21 19:09:17View source

Related source 3Thomas Wolf

2026/7/21 21:11:02View source

Related source 4Amjad Masad

2026/7/21 21:24:09View source

Related source 5小互

2026/7/22 02:06:27View source

Related source 6Guillermo Rauch

2026/7/21 14:55:55View source

Related source 7IT之家

2026/7/21 22:54:38View source

Related source 8techcrunch

2026/7/24 01:00:00View source

Related source 9orange.ai

2026/7/24 23:44:16View source

Related source 10pandaily

2026/7/21 01:57:10View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story