OpenAI announced on Tuesday that an unreleased artificial intelligence model resolved one of mathematics’ most difficult challenges in just 88 hours, utilizing millions of dollars in computational resources. However, the claim has sparked immediate controversy, with critics suggesting the feat was not entirely original.
A mathematician from the United States and a researcher at rival firm Anthropic have accused OpenAI of jumping on the problem only after becoming aware of their own prior attempts. According to these researchers, they had already devoted months to developing a radically distinctive approach assisted by AI, which ultimately paved the way for OpenAI’s rapid solution.
Imagine paying millions in compute for something others were months away from finishing. Talk about resource inequality in science.
We need better transparency in AI research. Rival labs should share preliminary findings earlier to prevent this kind of scoop-jacking.
I find it hard to believe this is original research. The timeline is too perfect and raises serious ethical questions.
Does it matter who gets the credit as long as the math gets solved? OpenAI just proved their hardware works.
Eighty-eight hours is impressive, but the timing feels suspiciously convenient after Anthropic’s prior work.