# Noam Brown 深度探讨多智能体与 AI 递归自我改进

- 来源：Noam Brown (@polynoamial)
- 发布时间：2026-09-18 00:20
- AIHOT 分数：44
- AIHOT 链接：https://aihot.news/items/cmu5qym1h0gu9roqochr9jnox
- 原文链接：https://x.com/polynoamial/status/2100620951822151719

## AI 摘要

OpenAI 的 Noam Brown 与 Dwarkesh Patel 联合发布了一期深度对话，主题涵盖多智能体、Navier-Stokes 方程，以及当前数学领域的爆发式进展对自动化 AI 研究意味着什么。两人还讨论了在启动递归自我改进（RSI）之前，如何判断模型是否真正对齐，以及思维链正在退化、内部/外部模型差距等话题。

## 正文

Happy to finally do a deep dive on multi-agent with @dwarkesh_sp! None of it would have happened without the great work on multi-agent from my @OpenAI teammates @kevinleestone, @mikegmalek, @eknight, @amuellerml, @zhangir_azerbay, @CheukHeiChu, and many others.

### 引用推文

> Dwarkesh Patel：New episode with @polynoamial We talk about multi-agent, Navier-Stokes, and what the current explosion of maths progress tells us about what happens once you au...