Dwarkesh Patel
2:37:16Rebuilding AlphaGo from Scratch: What Search, Self-Play, and Supervision Reveal About RL for LLMs
Eric Jang, most recently vice president of AI at 1X Technologies and earlier a senior research scientist at what is now Google DeepMind Robotics, spent part of a sabbatical rebuilding and hacking on A
2:13:48How Ancient DNA Revealed Accelerating Selection in the Bronze Age, and a New Model of Who the Neanderthals Were
For decades, the mainstream view in human evolution held that natural selection has been largely quiet in our species for the past several hundred thousand years. Geneticist David Reich of Harvard, sp
2:13:39How Frontier LLMs Are Trained and Served: Reiner Pope's Blackboard Lecture
Dwarkesh Patel's conversation with Reiner Pope, CEO of the chip startup MatX and previously a TPU architect at Google, was a blackboard lecture rather than a normal interview. The premise was that a f
1:43:10Jensen Huang on Nvidia's Moat: Supply Chains, TPUs, CUDA, and the Case for Selling Chips to China
In this long interview, Dwarkesh presses Nvidia CEO Jensen Huang on whether Nvidia's dominance can last. The questions cover whether Nvidia is really a software company that others manufacture for, wh
2:03:02Why Aliens Might Build a Different Tech Stack: Michael Nielsen on How Science Actually Progresses
Dwarkesh Patel opens this conversation with Michael Nielsen, a pioneer of quantum computing, co-author of the field's standard textbook, and a longtime advocate of open science, by setting aside most
1:23:43Terence Tao on Kepler, Verification, and Why AI Makes Mathematics Broader but Not Yet Deeper
In this conversation, Dwarkesh Patel asks Terence Tao how AI fits into the way science and mathematics make progress. They start with a historical case: how Kepler found the laws of planetary motion.
2:30:42Dylan Patel on Why Chips, Not Power, Will Cap AI Compute This Decade
In this long conversation, Dwarkesh Patel asks SemiAnalysis CEO Dylan Patel a series of questions about the AI compute buildout: what the hundreds of billions in hyperscaler CapEx is actually buying,
24:38Who Should AI Obey? Dwarkesh on the Anthropic–Pentagon Standoff
The Department of War has designated Anthropic a supply chain risk after the company refused to remove its red lines against using its models for mass surveillance and autonomous weapons. In this narr
2:02:19How Cosplaying Ancient Rome Led, Slowly, to Science: Ada Palmer on Renaissance Florence, Gutenberg, and Censorship
Ada Palmer is a Renaissance historian, novelist, and composer at the University of Chicago and the author of Inventing the Renaissance. In this conversation with Dwarkesh Patel she lays out a long cha
2:22:00Dario Amodei: "We Are Near the End of the Exponential"
Three years after their first conversation, Dwarkesh Patel asked Anthropic CEO Dario Amodei what had changed most. Amodei's answer was that the technology itself had not surprised him much. Models had
2:49:21Elon Musk on Orbital Data Centers, Grok's Values, Optimus, and the Limiting Factor
In a nearly three-hour conversation with Dwarkesh Patel and their co-host John, Elon Musk argued that the binding constraint on AI is no longer ideas or even chips. It is electricity. His answer is to
1:49:28Adam Marblestone on the Brain's Hidden Advantage: Reward Functions, Not Architecture
Dwarkesh Patel opens with what he calls his "million-dollar question": LLMs are trained on far more data than any human ever sees, yet they have only a fraction of human capabilities. How does the bra
12:33What Are We Scaling? Dwarkesh Patel on RL, Continual Learning, and Why AGI Isn't Imminent
This video is a narrated essay from Dwarkesh Patel's blog. Its central question is a tension the author sees in current AI forecasting. Many people expect AGI within a few years, and they are also ent
1:54:26Why Russia Lost the Cold War: Sarah Paine's Tour of the Competing Explanations
With the world possibly entering a second Cold War, historian Sarah Paine returns to the first one to ask why the Soviet Union lost it, and why participants on both sides thought it ended the way it d
1:35:39Ilya Sutskever: From the Age of Scaling Back to the Age of Research
In this conversation with host Dwarkesh, Ilya Sutskever, cofounder of Safe Superintelligence (SSI), argues that the recipe that drove AI progress for the past five years is reaching its limits. He say
1:28:17Satya Nadella on AGI, Fairwater, and Why Microsoft Won't Bet on a Single Model
Dwarkesh Patel and Dylan Patel, founder of SemiAnalysis, interviewed Microsoft CEO Satya Nadella in Atlanta after a tour of Microsoft's new Fairwater 2 datacenter. Their central question was how a com
1:30:59How Russia Repeatedly Derailed China's Rise: Sarah Paine on a Century of Russo-Chinese Relations
Many observers worry that China and Russia are forming a lasting partnership. Military historian Sarah Paine, whose dissertation traced the Russo-Chinese border from the Opium Wars to the 1920s detach
2:25:42Andrej Karpathy on Why Agents Will Take a Decade: Ghosts, Terrible RL, and the March of Nines
In this long conversation with Dwarkesh, Andrej Karpathy explains why he thinks current AI timelines are too optimistic, even though he calls himself very bullish on the technology. He calls this "the
1:20:28Nick Lane on Why Life May Be Chemically Inevitable but Complex Life Is Rare
Evolutionary biochemist Nick Lane of University College London explains life's four-billion-year history through energy flow. In this conversation with Dwarkesh Patel, Lane lays out a chain of reasoni
11:39Rethinking the Sutton Interview: Why Imitation Learning and RL May Not Be Opposites
After a widely discussed interview with Richard Sutton, Dwarkesh returned to the conversation to lay out what they now understand Sutton's worldview to be, and where they still disagree. Their central
