NewsletterUsersFeeds

Dwarkesh Podcast: Blog

All my posts

Sort by:
Pretraining progress is mostly coming from data
September 8th, 2026•dwarkesh.com
The Rise and Fall of Agent Civilizations
August 29th, 2026•dwarkesh.com
Why compute might get 10x+ more expensive in coming years
July 29th, 2026•dwarkesh.com
The Winning Essays for the Big Questions About AI
July 1st, 2026•dwarkesh.com
The sample efficiency black hole
June 8th, 2026•dwarkesh.com
The mistake of conflating intelligence and power
May 16th, 2026•dwarkesh.com
Notes on pretraining parallelisms and failed training runs.
May 16th, 2026•dwarkesh.com
RLVR might be disproportionately bad at science
May 16th, 2026•dwarkesh.com
What I've been thinking about this weekend - More open questions, intelligence vs power, the problem of verification in science, the parallel discovery of Darwinism
April 27th, 2026•dwarkesh.com
Blog prize for the big questions about AI
April 24th, 2026•dwarkesh.com
What I learned this week - Pretraining parallelisms, Can distillation be stopped, Mythos and the cybersecurity equilibrium, Pipeline RL, On why pretraining runs fails
April 15th, 2026•dwarkesh.com
Notes on Space GPUs
February 5th, 2026•dwarkesh.com
Hiring scouts to help me find guests
January 15th, 2026•dwarkesh.com
What I've been reading recently - Jan 10, 2026
January 10th, 2026•dwarkesh.com
Thoughts on AI progress (Dec 2025)
December 2nd, 2025•dwarkesh.com
Podcast Strategy Doc (December 2025)
December 1st, 2025•dwarkesh.com
RL is even more information inefficient than you thought
November 17th, 2025•dwarkesh.com
Thoughts on the AI buildout
October 22nd, 2025•dwarkesh.com
On The Vital Question by Nick Lane
October 1st, 2025•dwarkesh.com
On Kotkin's 2 volumes on Stalin - notes and questions
July 1st, 2025•dwarkesh.com