03 / Research & writing

Blogs.

Research notes and interactive explanations by Dane Malenfant on multi-agent learning, AI safety, and the mechanisms behind cooperative behaviour.

Research notes · Interactive explanations

Long-form notes that connect the questions behind my research to the experiments used to study them.

Query, team-success, and local-reward rates in the Dialogue Moral Hazard Game
Dialogue Moral Hazard / Read and play

· Interactive research note

Interactive Dialogue Moral Hazard

Why would an agent pay a private cost to uncover information that mainly protects somebody else? Read the mechanism, inspect the measurements, then play the Dialogue Moral Hazard Game yourself or let language agents play repeatedly.

  • Multi-agent language models
  • Moral hazard
  • Interactive experiment
Read and play