this should feel surprising. in some sense it's objectively very big news. but this was so clearly the trajectory we were on that it actually just feels like another normal day. life on the METR trendline.

@OpenAI

Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946.

For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids.

An OpenAI model has now disproved that… pic.twitter.com/j2g3Ze0zEG

at this point historical experiments like Talkie are unnecessary. we do not need to check whether LLMs merely "remix training data" vs make new discoveries by checking if it can rediscover relativity. here we are. a novel and contentful proof to a major outstanding conjecture!