OpenAI's AI claims major proofs in math and physics, awaits expert verification
Source published ·Modelwire updated
Original coverage: The Decoder ↗·How Modelwire adds context
The development
OpenAI's internal models have reportedly solved three open problems spanning mathematics, computer science, and physics, with one tied to a Millennium Prize Problem worth $1 million. The company has enlisted external validators, including a Nobel laureate, to verify the proofs. If the computer science result withstands scrutiny, it could invalidate hundreds of dependent research papers, signaling a potential inflection point in AI's capacity to generate novel scientific contributions rather than merely synthesizing existing knowledge. This moment tests whether frontier models can move from pattern recognition to genuine mathematical discovery.
Modelwire’s AI-generated summary of coverage from The Decoder.
Modelwire analysis
Skeptical readOur AI-generated reading of the wider context and the next developments to watch.
The proofs have not yet been verified. External validators are still reviewing them, which means the headline claim, that OpenAI's models solved a Millennium Prize Problem, is currently an assertion by the party with the most to gain from it being true.
This story is a direct continuation of the October 26 Verge piece on OpenAI announcing three science discoveries with third-party 'shepherds' embedded in the process. That coverage flagged the structural tension: as labs scale research output, independent verification becomes both more critical and more operationally complex. What the current story adds is the specific weight of the claims, a Millennium Prize Problem and a result that could invalidate hundreds of papers, which raises the stakes for that shepherd model considerably. Anthropic's formal verification work from October 23 (The Decoder) is worth holding alongside this: Claude's proof-checking covered existing code at scale, while OpenAI is asserting novel discovery, a meaningfully harder bar to clear.
Watch whether any of the named external validators, particularly the Nobel laureate, publish an independent written assessment within the next 60 days. A co-signed preprint carries different weight than a quoted endorsement in a press release.
This interpretation is generated from the summary above and available source metadata. Our methodology · Report an error
MentionsOpenAI · Millennium Prize Problem · Nobel laureate
How this coverage is produced
Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.
Modelwire summarizes, we don’t republish. The Decoder originally reported this story as “OpenAI asks outside experts, including a Nobel laureate, to check proofs its AI produced in math, computer science, and physics”. The full content lives on the-decoder.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.