OpenAI said on October 26, 2026, that an internal AI system produced a proof challenging the Exponential Time Hypothesis (ETH), a foundational assumption in theoretical computer science. The company published its argument and machine-checkable formalization while inviting independent researchers to examine the result. The claim is consequential, but expert interpretation is not finished.

Key Takeaways

• OpenAI says its system refuted ETH, a conjecture about the time needed to solve Boolean satisfiability problems.
• A Lean formalization is public; independent expert examination remains underway.
• The result does not establish that P equals NP or automatically speed up everyday software.

What Is the Exponential Time Hypothesis?

In plain terms, ETH holds that certain satisfiability problems cannot be solved in subexponential time as their input size grows. Such assumptions help computer scientists establish conditional limits on algorithms for scheduling, graph problems, databases and other demanding tasks. It is not the cryptocurrency Ether or the Ethereum network.

OpenAI’s October 26 announcement says its AI found a natural-language disproof on October 16 and formalized it in Lean on October 17. The company released the research manuscript and formal proof material so others can assess the claim.

Who Will Check the Result?

Russell Impagliazzo and Ramamohan Paturi of UC San Diego, alongside MIT’s Ryan Williams, will lead an external group investigating the argument and its implications. Impagliazzo and Paturi helped propose ETH. Mechanical proof verification checks formal logic; it does not replace expert explanation, scrutiny of assumptions or community consensus.

A Techmeme X post shared the announcement as the scientific community began discussing its significance.

A Larger Debate About AI-Generated Proofs

The same announcement describes claimed progress on the Kakeya conjecture and quantum field theory, with different public-release and review arrangements. BitcoinVersus.Tech previously covered OpenAI’s Navier–Stokes proof controversy and AI-assisted fusion-physics research. The site also documented Stanford’s virtual biotech with 37,000 AI scientist agents. Those stories illustrate why proving a statement and understanding its consequences are separate milestones.

The practical takeaway is not an immediate breakthrough for consumer computing. It is a potentially important theoretical result now entering public scrutiny, where reproducibility, independent review and clear explanations will determine its lasting significance.

BitcoinVersus.Tech Editor’s Note: We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. To support independent research, please donate 5,000 sats using the QR code below.

Scan this donation QR code from another screen, preferably a desktop or larger display, for easier camera recognition.