In a significant announcement, OpenAI has claimed that its new reasoning model has achieved a breakthrough by disproving a famous conjecture in geometry that has puzzled mathematicians since 1946. Originally posed by renowned mathematician Paul Erdős, this conjecture has influenced the understanding of geometric constructions for decades. If validated, OpenAI's assertion could redefine AI's role in addressing complex mathematical challenges.
This isn’t the first time OpenAI has made headlines with bold claims. Seven months ago, the company’s former VP, Kevil Weil, announced on social media that GPT-5 had solved ten previously unsolved Erdős problems and made progress on eleven others. Those claims were later debunked, revealing that GPT-5 only identified existing solutions from prior literature. The backlash from the mathematical community was swift, leading to a retraction of Weil's post.
In contrast, OpenAI seems to have adopted a more cautious approach this time. Along with the latest announcement, the company provided endorsements from respected mathematicians such as Noga Alon, Melanie Wood, and Thomas Bloom, who have confirmed the validity of the disproof. Bloom, who manages the Erdős Problems website, had previously criticized Weil's earlier claims as a “dramatic misrepresentation.”
OpenAI stated, “For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids. An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better.” This assertion indicates that the AI's reasoning capabilities can connect ideas across various fields, not just mathematics, with potential implications for biology, physics, engineering, and medicine.
The announcement has been met with cautious optimism within the AI and mathematical communities. AI's ability to handle complex reasoning and derive new insights could herald a new era for research and discovery. Thomas Bloom remarked, “AI is helping us to more fully explore the cathedral of mathematics we have built over the centuries. What other unseen wonders are waiting in the wings?”
If this claim withstands scrutiny, it could mark a significant moment for both AI and mathematics. By autonomously solving a prominent open problem, OpenAI is pushing the boundaries of AI's capabilities, potentially reshaping academic mathematics and the broader field of scientific inquiry.
As the implications of this development unfold, the future of AI in solving complex problems looks increasingly promising. Researchers and mathematicians will be watching closely to see how this breakthrough develops and what new avenues it might open for exploration in various scientific fields.
The stories that move AI & crypto markets — before the market reacts.
Free. 7am ET. Five stories. 62,400 readers.
