Fort Worth 24

collapse
Home / Daily News Analysis / Sam Altman sympathizes with mathematicians, but says OpenAI’s math onslaught won’t stop

Sam Altman sympathizes with mathematicians, but says OpenAI’s math onslaught won’t stop

Oct 31, 2026  Twila Rosenbaum  45 views
Sam Altman sympathizes with mathematicians, but says OpenAI’s math onslaught won’t stop

Key Facts

  • OpenAI CEO Sam Altman said he sympathizes with mathematicians trying to absorb a rapid stream of AI-generated results, but he does not intend to slow the company’s math research.
  • Altman argued that the public’s right to new scientific knowledge outweighs the disruption caused by OpenAI’s releases, while acknowledging the company must improve communication.
  • Researchers have criticized the pace of releases, citing inadequate attribution, little context, and insufficient verification. They have also accused OpenAI of withholding details about some claimed breakthroughs.
  • OpenAI has released hundreds of mathematical results this month alone and announced three major discoveries in mathematics, physics, and computer science. Details of a Yang-Mills result have not been made public.
  • Altman estimated OpenAI has spent hundreds of millions of dollars on math, including about $10 million or slightly more on a Navier-Stokes solution. He said the company holds many more undisclosed discoveries.

A Flood of Results Meets a Strained Field

After months of releasing a relentless stream of AI-generated mathematical results, OpenAI CEO Sam Altman said he sympathizes with researchers struggling to make sense of the deluge. But he made clear that the company has no intention of stopping—and suggested that many more discoveries remain in its stockpile.

“I think, on net, the world deserves to understand how the universe works,” Altman said during a podcast interview released Thursday. “But I do have empathy for the method.”

The remarks are among the first times Altman has directly addressed the impact of OpenAI’s math program on mathematicians. The company has produced a series of genuinely impressive results, but the rollouts have often been chaotic and poorly communicated, leaving researchers excited about AI’s potential for discovery while also bewildered and struggling to keep up.

Altman argued that the public’s right to new scientific knowledge “outweighs” the disruption caused by OpenAI’s releases. He acknowledged, however, that the company needs to improve how it communicates its findings. “I feel unequivocal on this point: Humanity deserves fundamental knowledge, and we should not be sitting on a major fundamental science discovery,” he said. “The world deserves that.”

Mathematicians have repeatedly criticized OpenAI’s announcements for inadequate attribution, little context, and insufficient verification. Researchers say the approach has left them doing painstaking work to check and interpret results while the company presses ahead with more. This month alone, OpenAI has released hundreds of mathematical results, often with brief writeups that researchers describe as poorly written and hard to follow. The volume has strained the field’s normal mechanisms for review, attribution, and collaboration.

The tension is not simply about volume. Mathematical results are typically judged through peer review, careful proof checking, and, increasingly, formal verification using proof assistants. AI-generated proofs can be difficult to evaluate because they may skip steps, combine known results in unfamiliar ways, or rely on assumptions that are not immediately visible. Even when a result is correct, the work of turning it into a communicable proof can be substantial. Researchers have said that OpenAI’s releases often shift that labor onto the community without providing enough documentation or support.

At the same time, the potential payoff is enormous. Mathematics is one of the areas where AI systems have shown notable progress, from solving olympiad-style problems to assisting with research-level conjectures. If AI can help identify new connections or generate candidate proofs, it could accelerate discovery in fields that depend on deep mathematical foundations, including physics, computer science, cryptography, and engineering. That promise helps explain why labs are willing to spend heavily on the problem.

A Gatekeeping Paradox

Yet OpenAI is now facing criticism for doing precisely what Altman says it should not do: sitting on potentially significant findings. Mathematicians and scientists have said they are frustrated by what they see as OpenAI’s gatekeeping of research, following the announcement of three major breakthroughs in mathematics, physics, and computer science. Details for one of the claimed results—involving Yang-Mills theory, the mathematical foundations of modern particle physics—have not been released publicly.

OpenAI says the delay is due to the recommendations of a hand-picked group of experts it consulted on how best to release the result. The practice runs directly counter to the advice of a previous group of mathematicians the company consulted, but forms part of what the company describes as a new approach to communicating its findings. The contradiction has sharpened concerns that OpenAI is trying to control the narrative around discoveries that, by its own public-interest argument, should be shared openly.

The episode illustrates a broader problem in AI-driven science. When a lab announces a result before the underlying proof, data, or reasoning are available, the scientific community cannot independently assess it. That can fuel hype, confuse the public, and make it harder for researchers to build on the work. If a lab withholds details to manage scrutiny or protect priority, it can also undermine trust. The result is a communication model that satisfies neither the demand for openness nor the need for careful verification.

The Cost and Scale of AI Mathematics

On the podcast, Altman gave a rare glimpse of the resources OpenAI is devoting to mathematics. He estimated that the company has spent in the order of hundreds of millions of dollars tackling mathematical problems, including around $10 million—“maybe a little more”—on its Navier-Stokes solution, one of mathematics’s legendary Millennium Prize problems.

Navier-Stokes is central to fluid dynamics and is one of the seven Millennium Prize problems, a set of exceptionally difficult questions that carry a $1 million prize for a correct solution. A credible AI-assisted advance would be a major scientific event. But the disclosure of the spending also highlights the imbalance between well-funded AI labs and academic researchers. Mathematicians have raised concerns that they may be priced out of their own fields as compute, data, and engineering talent become decisive advantages in tackling hard problems.

The scale of spending also changes expectations. When a company invests hundreds of millions of dollars in a research area, it creates pressure to show results—and to frame those results as breakthroughs. That pressure can conflict with the slower, more collaborative norms of mathematical research, where priority is usually established through publication, peer review, and the gradual acceptance of a proof by experts.

Internal Experiments and Undisclosed Results

Altman also acknowledged that OpenAI’s pursuit of mathematical problems did “get out of control” for a short period of time, but said that was deliberate and necessary to probe the limits of its AI models. He described an internal Slack channel called “Model Name-Hold My Beer,” where employees “were just throwing that at the hardest problems they could” to see how the models would perform.

The anecdote offers a window into the experimental culture inside OpenAI, where researchers can test models against difficult problems at speed and scale. But it also reinforces the sense that the company’s math program has been driven partly by curiosity and internal competition, with public communication following later—or not at all.

Altman hinted that OpenAI is sitting on many more mathematical discoveries than it has made public. Asked whether the company has more than the hundreds of proofs it has already released, Altman replied: “We do. Of course.”

That admission sits awkwardly beside his argument that humanity deserves fundamental knowledge and that OpenAI should not sit on major discoveries. The company has not explained how it decides which results to release, how it verifies them before announcement, or how it balances priority, safety, and public access. For now, mathematicians hoping for a breather appear unlikely to get one. OpenAI has not publicly said when the withheld Yang-Mills details will be released, even as Altman says more mathematical discoveries are already in hand.


Source: The Verge News


Share:

Your experience on this site will be improved by allowing cookies Cookie Policy