OpenAI says its AI solved three open problems and asks outside experts to check
OpenAI says its internal AI models solved three major open questions in mathematics, computer science and physics. Outside experts, including a Nobel laureate, are now checking whether the proofs hold up.
OpenAI says its internal AI models have cracked three major open questions in mathematics, computer science and physics. Before the results can be treated as established, outside experts, including a Nobel laureate, are checking whether the proofs hold up.
One of the three questions is part of a Millennium Prize Problem, a set of seven famous mathematical challenges that each carry a prize of $1 million. Whether the arguments produced by the AI are sound is therefore likely to draw attention beyond the research community.
The consequences would be especially far-reaching in computer science. If the proof in that field holds on its own, hundreds of research papers would lose their foundation, because many publications rest on assumptions that would then no longer be valid.
The use of AI-generated proofs is contested in the scientific community. A mathematical proof is accepted only once specialists can follow and verify it. By handing the verification to external experts, OpenAI is seeking exactly that kind of recognition. It is not yet known when the review will be complete or what it will find.
8
