Noam Brown

Person

Heard in 7 episodes across 6 shows since Aug 2026

Noam Brown is a research scientist at OpenAI focused on multi-step reasoning, self-play, and multi-agent artificial intelligence. He co-created Libratus and Pluribus, superhuman poker-playing AIs, and CICERO, a Diplomacy-playing AI.

Discussed inAI & Tech56%Business44%

Episodes

across 6 shows

This week

episodes on 0 shows

Guest appearances

Share of the conversation

Mentions of Noam Brown per 1,000 mentions of anything, 7-day rolling, across the podcasts PodLume Trends follows.

Show the data

Day	Share (7-day)	Mentions
Aug 6	0.22‰	2
Aug 27	0.03‰	1
Sep 17	0.03‰	1
Sep 21	0.05‰	1
Sep 26	0.08‰	2
Sep 28	0.16‰	4
Oct 3	0.14‰	1

The full trend: rank, tone and why it is trending →

What Noam Brown has said on podcasts

52 statements

on AI reasoning modelsPositiveSep 17, 2026· Dwarkesh Podcast
More test-time compute improves reasoning-model benchmark performance.
“when you plot the performance of these reasoning models with test time compute on the x-axis and performance on basically any reasoning benchmark on the y-axis, you see a very clear pattern where the longer these models take to think about their answer, the better they do”
Listen at 0:50
Open the episode · Noam Brown – Agent swarms, alignment, & recursive self-improvement
on Multi-agent AI systemsPositiveSep 17, 2026· Dwarkesh Podcast
Multi-agent systems scale test-time compute through parallelization.
“multi-agent is a way of scaling test-time compute in parallel instead of purely serial”
Listen at 1:50
Open the episode · Noam Brown – Agent swarms, alignment, & recursive self-improvement
on Multi-agent AI systemsMixedSep 17, 2026· Dwarkesh Podcast
Parallel multi-agent scaling is effective but less efficient than single-agent reasoning.
“it is less efficient because it doesn't have— it's not like a single agent has all the context to itself, but it is a very effective way of scaling test-time compute if it's done well”
Listen at 1:55
Open the episode · Noam Brown – Agent swarms, alignment, & recursive self-improvement