Demis Hassabis, Andrej Karpathy, and Elon Musk discussed using the game Diplomacy to test AI. One AI researcher took them up on it and built a new game called "AI Diplomacy." He found that OpenAI's o3 ...

Earlier this year, some of the world's leading AI minds were chatting on X, as they do, about how to compare the capabilities of large language models. Andrej Karpathy, one of the cofounders of OpenAI ...