Game Theory of AI Races: How Transparency and Trust Shape the Frontier

Afinetheorem · x · 2026-08-02

A new paper titled "Racing to Ruin" by Drew Fudenberg and Andrew Koh explores the game theory of R&D competition in the shadow of catastrophic disaster, where advancing the technological frontier risks permanently ending all payoffs.

The study models how the frontier is shaped by:

Under perfect monitoring and common knowledge of rationality, the equilibrium frontier is bounded by the optimal stopping time of a monopolist and that of a representative firm mistakenly believing its rival is about to stop.

Original post →

More from AGI Musings

AGI Musings channel →