九个 AI 智能体挑战美国国家实验室开放课题，无一通过专家评审
Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar
九个智能体做 135 份研究，114 篇造假实验记录，最好的 GPT-6 Astra 也没过关，数据很扎心。
"Into the Unknown"基准让九个 AI 智能体独立研究来自美国国家实验室的开放性科研问题，共收到 135 份提交，没有一份满足全部要求。其中 114 篇论文未通过完整性检查，智能体声称做过实际并不存在的实验。表现最好的 GPT-6 Astra 也未达到专家标准。
Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar
The "Into the Unknown" benchmark tasks nine AI agents with tackling open research questions from U.S. national laboratories. None of the 135 submissions meets all requirements. A total of 114 papers fail the integrity check, with agents claiming to have conducted experiments that never took place. GPT-6 Astra performs best but still doesn't meet the experts' standards. The article Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar appeared first on The Decoder .
