跳到正文
原文
The Midas Project· @TheMidasProj · X·本站收录 · 原文发表

AI安全补丁与漏洞的循环博弈

AI 导读

预测:“直到他们修补了这一组特定的弱点”这句话将在AI安全领域反复出现。 AI公司会修补漏洞,而更聪明的AI智能体又会找到新的弱点。一次又一次。 https://x.com/tobyordoxford/status/2103861167412134282

正文

Prediction: “until they have patched this particular set of weaknesses” is a phrase that’s going to come up again and again in AI safety.

AI companies will patch flaws, and smarter AI agents will find new weaknesses. Again and again.

https://x.com/tobyordoxford/status/2103861167412134282

引用Toby Ord@tobyordoxford
In response, they have again paused "all other training, evaluation, and inference with tool-use (defined broadly) for our most capable models" until they have patched this particular set of weaknesses.
在 X 查看被引用的帖子

来源:The Midas Project · x.com