OpenAI model may have completed half of the million-dollar Millennium Prize Yang-Mills problem proof

November 12, according to MTS post, OpenAI's artificial intelligence model has proven half of a million-dollar Millennium Prize problem in the field of physics, and is supported by a review panel led by a Nobel laureate. MTS stated that the problem is the Yang-Mills problem. OpenAI first announced the relevant proof on October 26 and stated that its model completed the work in September.

According to MTS, the review panel is chaired by Nobel laureate David Gross and includes Arthur Jaffe, who co-authored the official formulation of the Millennium Prize problem. The relevant proof has also been formalized in Lean, a computer language that can be verified line by line. The Yang-Mills problem is one of the seven Millennium Prize Problems established in 2000, each with a $1 million prize; the only one solved so far was completed by Grigori Perelman, but he refused to accept the prize.

MTS pointed out that to win the prize, it is still necessary to prove the mass gap and satisfy the Clay Mathematics Institute's rules on publication and a two-year recognition period. X user Theo Jaffee stated that physicists have been able to handle quantum field theory and quantum Yang-Mills theory at the empirical level, but have always lacked a mathematical formalization of why this theory holds.