Event ·Multi-source

GPT-5.6 解决概率论中 Feige 1/e 猜想

OpenAI 联合创始人 Greg Brockman 公布 GPT-5.6 成功解决了概率论中长期未解决的 Feige 1/e 猜想。该猜想涉及独立非负随机变量 X_i（期望 ≤1）的部分和 S_n，断言 P(S_n ≤ E[S_n] + 1) ≥ 1/e。Brockman 的友人 Zhengqing 使用 GPT-5.6 完成了这一证明，该问题曾是他们午餐时反复讨论的难题。

Current read

GPT-5.6 刚又搞定一个数学难题，这次是概率论里老牌的 Feige 猜想，你可以看看 AI 证明数学定理有多猛。

11 sources67° AI heatLast updated 2026/7/29 03:36:57

Evidence

11 sources

Primary sourceGreg Brockman

2026/7/29 03:36:57View source

Related source 1lmarena.ai

2026/7/27 21:06:42View source

Related source 2Amjad Masad

2026/7/28 07:09:33View source

Related source 3OpenAI Blog

2026/7/29 00:00:00View source

Related source 4Simon Willison

2026/7/30 01:34:49View source

Related source 5Fireworks AI

2026/7/27 15:30:35View source

Related source 6@OpenAIDevs

2026/7/29 21:33:09View source

Related source 7SuperTechFans

2026/7/29 23:49:57View source

Related source 8Sam Altman

2026/7/30 00:41:57View source

Related source 9小互

2026/7/30 04:12:00View source

Related source 10Decoder

2026/7/30 07:40:11View 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