Subvolts take DeepMind's AI game mastery: AlphaGo and AlphaZero's journey from games to broader AI insights. Aug 20, 2019 - Developers
- Learn about reinforcement learning techniques used to train game playing AI like AlphaGo and AlphaZero.
- Business
- Understand how complex problem solving in games can translate to broader AI applications and strategic advantages.
- Everyday users
- Discover how AI achieved superhuman performance in games and what it means for the future of artificial intelligence.
This episode features key researchers discussing significant AI milestones in game playing, offering deep insights into their development.
Go to Zero
Audio streams from Hannah Fry. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Google DeepMind's podcast "Go to Zero" features Lead Researcher David Silver discussing AlphaGo and AlphaZero. The episode explores how AI learned games like Go, chess, and shogi, surpassing human capabilities. Chess Grandmaster Matthew Sadler and Natasha Regan, authors of "Game Changer," also join to share insights on AlphaZero's unique gameplay and its implications.
This episode explains how computers learned to play complex games like Go and chess better than humans. Researchers discuss how these games helped build smarter AI, showing how learning in one area can lead to advances in others.
This episode offers a look inside DeepMind's groundbreaking AI research, particularly its success in game playing with AlphaGo and AlphaZero. It's for anyone interested in the evolution of AI, game theory, and how complex systems learn. The discussion connects to current frontier lab efforts in developing more capable and general AI systems, showing how early research in constrained environments like games paved the way for broader AI advancements.
- HostHannah Fry
- GuestDavid SilverLead Researcher, Google DeepMind
- GuestMatthew SadlerChess Grandmaster
- GuestNatasha ReganWomen's International Master
- GuestDemis HassabisCEO, Google DeepMind
- GuestMatt BotvinickDirector of Neuroscience Research, Google DeepMind
- Episode Title: Go to Zero
- Show: Google DeepMind: The Podcast
- Length: 26 minutes
- AlphaGo Milestone: Defeated professional human player in March 2016
- AlphaZero Capabilities: Taught itself chess, Go, and shogi
In March 2016, more than 200 million people watched AlphaGo become first computer program to defeat a professional human player at the game of Go, a milestone in AI research that was considered to be a decade ahead of its time. Since then the team has continued to develop the system and recently unveiled AlphaZero: a program that has taught itself how to play chess, Go, and shogi. Hannah explores the inside story of both with Lead Researcher David Silver and finds out why games are a useful proving ground for AI researchers. She also meets Chess Grandmaster Matthew Sadler and women’s international master Natasha Regan, who have written a book on AlphaZero and its unique gameplay.
If you have a question or feedback on the series, message us on Twitter (@DeepMind using the hashtag #DMpodcast) or email us at podcast@deepmind.com.
Further reading
- AlphaGo the documentary
- The Surrounding Game: Documentary about the ancient game of Go
- DeepMind website: AlphaGo
- Garry Kasparov: Deep Thinking
- AI: More than Human - Exhibition at the Barbican Centre, 2019 and online exhibit
- DeepMind blog: AlphaZero: Shedding new light on chess, shogi, and Go
- Matthew Sadler and Natasha Regan: Game Changer - a book about chess and AI
- WIRED: What the AI behind AlphaGo can teach us about being human
Interviewees: DeepMind CEO Demis Hassabis, Matthew Sadler, chess Grandmaster; Lead Researcher David Silver, Matt Botvinick, Director of Neuroscience Research; and Natasha Regan, women’s international chess master.
Credits:
Presenter: Hannah Fry
Editor: David Prest
Senior Producer: Louisa Field
Producers: Amy Racs, Dan Hardoon
Binaural Sound: Lucinda Mason-Brown
Music composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)
Commissioned by DeepMind
Please leave us a review on Spotify or Apple Podcasts if you enjoyed this episode. We always want to hear from our audience whether that's in the form of feedback, new idea or a guest recommendation!
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
44 minCan AI help us better predict the weather?
WithPeter Battaglia
45 minThe mathematics of AI uncertainty
WithZoubin Ghahramani
53 minSame guests, other shows
1 h 08David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
WithDavid Silver, Lasse Espeholt
1 h 35Demis Hassabis on AGI, Robots Scale Production, and Elon’s $1T Mars-Shot Comp | EP #253
WithDemis Hassabis, Steven Kotler, Salim Ismail, Dave Blundin
27 min
59 minQuestions people ask
- What is AlphaGo and AlphaZero?
- AlphaGo was the first computer program to defeat a human professional at the game of Go. AlphaZero is a successor that taught itself chess, Go, and shogi.
- Why are games useful for AI research?
- Games provide a useful proving ground for AI researchers to develop and test advanced systems, pushing the boundaries of artificial intelligence.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Hannah Fry.