AlphaGo

AlphaGo is Google DeepMind’s Go-playing AI system, used in ^[src-063] as an example of learning a model that guides search through an otherwise enormous combinatorial space.

Key facts

• Type: Game-playing AI system
• Maker: Google DeepMind
• Source role: Evidence for the Learnable Natural Systems and guided-search framing ^[src-063]
• Hassabis groups AlphaGo with AlphaFold as an “Alpha” project that makes a huge search problem tractable by building a useful model of the space ^[src-063].
• The episode treats AlphaGo’s Lee Sedol match as an example of AI inspiring human creativity rather than only replacing human judgment ^[src-063].

What it adds

In this source, AlphaGo is less a standalone game milestone than a template for the DeepMind method: learn structure, search intelligently, and uncover surprising moves or solutions in a space humans cannot exhaustively enumerate ^[src-063].

Related entities

• Demis Hassabis
• Google DeepMind
• AlphaFold

Related concepts

• Learnable Natural Systems
• World Models
• AI For Science

Source references

• ^[src-063] Lex Fridman – “Demis Hassabis: Future of AI, Simulating Reality, Physics and Video Games | Lex Fridman Podcast #475” (2025-07-23)