Researcher Plinz: Everyone Predicting Hard Limits on LLM Abilities Ended Up Wrong

burny_tech · x · 2026-09-28

AI researcher Plinz tweeted that LLMs are improving very rapidly. Everyone who predicted hard limits on any specific ability has ended up with egg on their face: coding has gotten noticeably better in just the last month, and math abilities — almost nonexistent a couple of years ago — have advanced dramatically. The post pushes back against the common claim that LLMs face insurmountable capability ceilings.

Related event: Researcher Plinz: LLMs Have No Hard Capability Ceiling(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →