This isn't getting any attention, but Anthropic claims Claude independently proved that the Exponential Time Hypothesis was false back on October 16th, and had been working on a cleaned-up draft.
dooglius · 2026-10-27 16:59:58 UTC
It's a bit sus that the time complexity and also the fact it solves the counting version of the problem, match exactly the openAI result
QuesnayJr · 2026-10-27 17:02:19 UTC
It is weird. We'll have to wait and see if they are the same algorithm or not.
commandersaki · 2026-10-27 15:45:08 UTC
Is it 2^O(n) or 2^o(n)? (Edit: It's the latter according to the site, HN possibly mangled the title.)
Comments