Event ·Single source

GDPval 后继基准发布，Ethan Mollick 指出缺少人类基线

GDPval 的后继基准上线，用于评估 AI 完成真实工作任务的表现。Ethan Mollick 认为该基准仍需像 GDPval 一样补充人类表现基线。他提到 GPT-6 Pro 在评分中发现了若干错误，且 0% 的得分掩盖了不同模型在经济意义上的成功率差异。

Current read

Ethan Mollick 聊了 GDPval 的后继基准，还发现 GPT-6 Pro 能挑出评分错误，做模型评测的可以看看

2 sources47° AI heatLast updated 2026/11/2 06:30:33

Evidence

2 sources

Primary sourceEthan Mollick

2026/11/2 06:30:33View source

Related source 1Jerry Liu

2026/11/4 01:14:06View 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