The mathematics result is a proof of Sogge's local smoothing conjecture for the Euclidean wave equation in every spatial dimension of at least two — previously proved in only three dimensions.

The computer science paper claims the Exponential Time Hypothesis is false, and the physics work is described as a rigorous existence proof for four-dimensional quantum field theories including quantum chromodynamics. OpenAI says it will claim no academic credit for the discoveries.

The boldest claim directly overturns a foundation of theoretical computer science.

MIT's Ryan Williams, a complexity expert, told Al Jazeera that a confirmed refutation would be a major development, noting the proof "proposes a mysterious new method to more efficiently solve a variety of fundamental problems."

Engadget reports the paper gives a deterministic algorithm counting satisfying assignments of fixed-width CNF formulas in subexponential time and polynomial space, and that OpenAI says it was verified in the Lean proof assistant.

OpenAI itself frames the manuscripts as "a form of raw research data" — inputs for experts to test rather than finished verdicts.

Two manuscripts were added to GitHub: the local smoothing proof, which OpenAI says also yields the Kakeya maximal and dimension conjectures and Stein's spherical restriction estimates, and the ETH refutation with its Lean proof.

The Verge reports that Lean verification confirms a formal argument is valid but cannot confirm the formal statement matches the informal claim.

None of the three results has been accepted by the scientific community as settled knowledge; each remains, for now, an OpenAI claim awaiting independent verification.

The release experiments with three strategies. For computer science, a group led by Impagliazzo, Paturi and Williams — including the hypothesis's original proposers — released the materials publicly and an open letter asking the community to set norms for AI-generated breakthroughs.

For physics, OpenAI deferred to an external group led by David Gross, including Michael Douglas, Arthur Jaffe, Igor Klebanov and Nikita Nekrasov, which has not yet published its interpretation.

The Verge notes OpenAI's own AGMAI panel of math advisers had just recommended that AI companies avoid designating particular people to receive output first, citing unequal access, borrowed reputation and biased selection.

If the claims hold up, the consequences are sweeping. A full local smoothing proof would resolve problems connected to the Kakeya conjecture — for which NYU's Hong Wang won a 2026 Fields Medal for work in three dimensions, according to OpenAI's announcement as reported by Firstpost.

A false ETH would undermine conditional lower bounds across fine-grained complexity, though Williams told Al Jazeera short-term real-world applications are unlikely even as future benefits for society could follow.

Rochester mathematician Alex Iosevich called the harmonic analysis development "historic," telling Al Jazeera he expects theoretical mathematics to "change drastically" while acknowledging a "twinge of sadness" among colleagues.

Verification is the immediate task.

The computer science group has posted an open letter asking three broad questions: what expectations should govern early access to AI-generated results, how credit should recognize both discoveries and the labor of checking and explaining them, and how to support students and researchers as methods change.

The physics group has offered an email contact — YANGMILLS4D@gmail.com, per Engadget — for questions while it works on its own writeup. OpenAI says participation in the collaborations is voluntary and unpaid, and welcomes feedback on how to broaden participation.

The pattern follows Anthropic's contested claim last month of a new CRISPR-like enzyme, which scientists greeted with mixed reactions, per Al Jazeera.