Speculation: OpenAI quietly throwing massive compute at remaining Millennium Problems

haider1 · x · 2026-11-22

haider1 speculates (unverified) that OpenAI is already training/running an internal model with massive compute on the remaining Millennium Prize Problems, since it can't afford to let Anthropic lead in math. He argues OpenAI likely has a model equally capable as Claude but far less trustworthy, and that trustworthiness is its biggest bottleneck.

Original post →

More from AGI Musings

AGI Musings channel →