ETH conjecture disproved by OpenAI AND Anthropic on the same day?!

This is truly a gargantuan breakthrough in computer science. The exponential time hypothesis is false. This renders immediately obsolete hundreds or probably thousands of papers in computer science which assumed that this hypothesis was true. In fact, many people keep saying AI just solves math problems. It doesn't ever open new math problems. But today, OpenAI

opened up many new math problems because all of the results which depended on this hypothesis are now wide open. But perhaps the most remarkable thing is that the breakthrough was made simultaneously by both Open AI and Anthropic. They both discovered this breakthrough on the same day, October 16th, apparently completely independently of each other. In the

announcement from OpenAI, we read, "Our AI found its natural language disproved for the exponential time hypothesis on October 16th, 2026, and then subsequently we formalized the result in lean on October 17th, 2026. Mere hours after open AI posted this announcement, Levent Alpagave posted on Twitter, the core algorithm was found on October 16th by an internal Claude

research model with 14 million output tokens and no human input. Anthropic researchers worked with Claude to improve the presentation resulting in this paper and he links to a paper from Anthropic. Notice October 16th for Anthropic, October 16th for OpenAI. They both got the same breakthrough on the exact same day. What are the odds? This is such an

incredible coincidence. Or perhaps it's more than a coincidence. I'm sure the conspiracy theorists will have an absolute field day with this one. I'm Dr. Samuel Allen Alexander, mathematician and philosopher still here. Over on Twitter, user Chopper89 posted this helpful table comparing and contrasting the two breakthroughs. Now, both of them address the they they refute the exponential