Thoughts on building AI agents that reason, compete, and cooperate — from game-theoretic LLMs to multi-agent worlds. By Seth Karten, PhD candidate at Princeton.