AI researchers debate how close we are to recursive self-improvement
Dwarkesh Podcast

AI researchers debate how close we are to recursive self-improvement

B2中高级2026年9月11日1小时 37分钟
  • 访谈
  • 科技

New episode with John Schulman, Beren Millidge and Charlie O’Neill. I got together with some of the most insightful AI researchers I know who are at the openish companies, because I wanted to hear the details of what's actually happening at the frontier and what comes next. Watch on YouTube; read the transcript. Sponsors * Antithesis helps you trust your code. As agents generate more and more of your software, the bottleneck shifts from your engineers actually writing code to verifying it. Antithesis does that testing for you. Ron Minsky, who co-leads Jane St...

播放本集

打开播放器:完整文字稿,每个单词都可点击,附翻译和单词卡。

试一试:点任意单词

来自本集开头的真实句子,每个单词都标出了难度级别。

第 1 句,共 12 句 · 0:00

,

from whom I learn a lot every time we talk,

点一个单词,用你的语言查看释义。

这些单词你认识吗?

本集中 8 个较难的单词。点一个查看释义,然后标记一下。

标记几个单词,我们会告诉你这一集是否适合你的水平。

这一集适合你吗?

B2

B2 · 中高级

能看懂电视新闻就会觉得轻松

A1 22%A2 30%B1 25%B2 19%C1 5%
  • 76% 的单词在 B1 或更简单的水平。
  • 每分钟 197 词:自然语速,偏快。
  • 整集共有 392 个较难的单词(B2 及以上)。

文字稿

本集开头部分。完整文字稿在播放器中。

0:00

Today, I'm chatting with three of my AI researcher friends from whom I learn a lot every time we talk, and who also happen to be at somewhat open-ish labs and companies,

0:09

so you guys can actually say things on the record. I'm joined by Baron Milish, who is the CTO of Zephyr, which is developing open source models. John Shulman, who is the Chief Scientist at Thinking Machines,

0:19

previously the co-founder of OpenAI, led the R-Litch F-Work that led to Chetchee PT, and Charlie O'Neill, who is Head of Model Training at Base 10.

以上是开头部分。 本集其余部分(97 分钟)在播放器中:每个单词都可点击,附翻译、生词收藏和单词卡。

在播放器中继续

由音频自动生成,可能存在错误。 音频版权归其发布者所有,并从其订阅源播放。版权所有者可申请删除 — 版权与删除政策