150+ Mathematicians Challenge AI Problem-Solving Claims

A coalition of more than 150 mathematics experts has issued a stark warning to policymakers: don't believe the hype about artificial intelligence solving complex mathematical problems. The Leiden Declaration on AI and Mathematics represents the strongest public pushback yet against recent claims from companies like OpenAI.

The controversy

The declaration comes after OpenAI claimed its AI systems had autonomously solved prominent mathematical problems, including disproving an 80-year-old conjecture devised by Hungarian mathematician Paul Erdős. The company called it "the first time AI has autonomously solved a prominent open problem central to a field of mathematics."

But the mathematics community isn't convinced. The 11-page declaration, signed by experts worldwide, explicitly warns that "there is currently a strong commercial incentive on the part of the technology industry to overstate the capabilities of their products."

Why it matters

Mathematical research forms the foundation for advances across science and technology. If AI systems produce plausible but incorrect proofs that enter the literature, researchers building on that work could waste years pursuing dead ends. The declaration highlights a critical tension between commercial AI hype and scientific rigor that extends far beyond mathematics into peer review, research integrity, and technology policy.

The reliability problem

Leslie Ann Goldberg, head of computer science at the University of Oxford and a declaration signatory, emphasized the core concern: "Current automated techniques can produce plausible but unreliable (or even incorrect) arguments which are difficult to distinguish from correct mathematical proofs."

This poses a fundamental challenge for mathematical research, which builds incrementally on previous results. If researchers cannot trust the correctness of published work, the entire edifice becomes unstable.

Broader concerns

The declaration also addresses systemic pressures facing academics. Many mathematicians face precarious employment and funding shortages while AI companies offer "lucrative jobs, monetary rewards, computing resources, and intellectually stimulating opportunities." This creates incentives to endorse AI capabilities regardless of actual performance.

The document raises additional concerns beyond mathematics, including AI's role in military surveillance, misinformation, democratic undermining, and environmental costs.

Training without consent

Rodrigo Ochigame, a Leiden University anthropologist of AI who helped draft the declaration, highlighted another troubling dimension: "Mathematicians who never intended to contribute to AI development are having their work used for this purpose without their consent."

The path forward

International Mathematical Union vice president Ulrike Tillmann stated that "the future of mathematical research must be guided by human judgment, fair and transparent practices, and the shared values of the global mathematical community."

The declaration advises policymakers to "consult with experts, including mathematicians, in forming policy decisions rather than relying on press releases or popular reporting of mathematical results."

These details were first reported by AI Watch and Scientific American.