Skip to content

Comment on AI Built a Nuke and Still Lost

Comments

LLMs are really bad at abstract strategy games like chess, go or civilization. Their ability to excel at broad reasoning is what is limiting them in games that have narrow rule-sets but steep learning curve.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.