Glossary · Term

self-play

← all terms

Definition

Plain language

Training an AI by having copies of it challenge each other, instead of learning from human examples.

As stated in the literature

A training paradigm where a model improves by generating its own challenges and solutions — e.g., one instance proposing tasks and another solving them; effective but prone to curriculum drift and regression without targeting toward the real objective.

Why it matters: It lets a model keep improving from its own activity rather than scarce human data, though it can drift off course unless steered toward the real goal.

For example, one copy of a model invents practice problems while another copy tries to solve them, and both get better from the back-and-forth without any human-written examples.

Heard on the show

“And as post-training pushes models further from human data — reinforcement learning on verifiable rewards, multi-agent self-play — models will correctly infer that humans are dissimilar.”
Episode 234 — Two Copies of Gemini Cooperated in a Game Where Betrayal Always Pays

Mentioned in 5 episodes

  1. 234
    Two Copies of Gemini Cooperated in a Game Where Betrayal Always Pays
  2. 207
    An AI Graded Its Own Math Test 94 Percent — It Actually Scored 20
  3. 206
    How Four-Second Clips Become Hours of Playable AI Soccer
  4. 126
    How Coding Agents Can Mine Their Own Failures Into a Self-Targeting Curriculum
  5. 018
    Language Models Compute the Rational Move, Then Override It

Related terms