OpenAI announces three discoveries in mathematics, computer science, and physics, and brings in external experts to vet them
OpenAI announced three "major" discoveries in mathematics, computer science, and physics. In computer science, it claims to have proven that ETH (the Exponential Time Hypothesis), long widely believed to be true, is in fact false, and the related AI output and Lean proof have been published on GitHub. The physics result has not yet been fully disclosed; OpenAI describes it as a rigorous existence proof for an interacting quantum field theory in four dimensions containing gauge fields and fermions, covering a unified completion of QCD and the Standard Model, and says David Gross and others will lead the vetting. In mathematics, its model produced a proof of the local smoothness conjecture in arbitrary dimensions, and thereby resolved the Kakeya conjecture and the Stein restriction conjecture in all dimensions; OpenAI also published an open letter soliciting opinions on how to test these conclusions.
Why read this
SelectedAI score 80
OpenAI announces three scientific discoveries and brings in external experts to vet them, so readers can understand the debate over verification and publication standards for AI-produced results.
Topics
Body
Less than a month after its mass release of hundreds of manuscripts that it said contained solutions to long-standing mathematics questions, OpenAI is back with three more “major” discoveries in math, computer science, and physics . What’s different this time is the choice to have external experts that OpenAI describes as “shepherds” guiding the results in computer science and physics, while for the math problem it says that they advised it to release the AI output for researchers to work through their own explanations.
This approach could have its own issues, the company admits, and the AGMAI panel of math advisers OpenAI put together just said yesterday that “we recommend that AI companies do not adopt” the practices of designating certain people to receive the output first, citing unequal access, borrowed reputation, and biased selection as a few of the issues. Instead, they recommended that in the future, AI labs release output on scholarly repositories, with Lean-formalized proofs, info on how the problems were chosen, funding for people working on understanding the results, and other information.
The physics information has not been released in full, but is a “mathematically rigorous existence proof of certain interacting quantum field theories containing gauge fields and fermions in four dimensions, including Quantum Chromodynamics (QCD) and a unified completion of the Standard Model of particle physics.” According to OpenAI, “David Gross will lead a group including Michael Douglas, Arthur Jaffe, Igor Klebanov, and Nikita Nekrasov that will shepherd this work.”
The computer science collaboration is said to prove that the Exponential Time Hypothesis (ETH) , which has been widely believed to be true, is actually false. OpenAI says verification of the result will be guided by a group of three experts that includes UC San Diego’s Russell Impagliazzo and Ramamohan Paturi, who were part of the group that originally proposed it in 1999 , as well as MIT professor Ryan Williams . The group decided to release the AI output write-up and Lean proof on GitHub today .
They also released an open letter to the community asking for input on how to examine the claim, and about setting norms for dealing with AI-generated breakthroughs:
There’s similar documentation for the math breakthrough, where its model “produced a proof of the local smoothing conjecture in every dimension and therefore also resolve the Kakeya conjecture and Stein’s restriction conjecture in all dimensions.” Just a few months ago, NYU professor Hong Wang was awarded a prestigious Fields Medal by the International Mathematical Union for resolving the Kakeya conjecture in three dimensions.