Who signed the AI and mathematics declaration?

The declaration carries 25 initial signatories, all of them Fields Medallists. Terence Tao, himself a Fields Medallist, posted it on his blog "What's new" on 11 September 2026 and hosts the full text at mathandai.org. Tao's post identifies him as a signatory but does not name any single author of the declaration.

Does the declaration name any AI company?

Terence Tao's post points readers to a recent Economist article describing OpenAI's methods, and OpenAI is the company named there. The declaration's argument is aimed at the wider industry practice of using open mathematical problems as a capability benchmark, not at one lab's individual products.

Why do the signatories raise plagiarism and attribution?

The declaration says AI results are "announced in a rush, leaving no time for a proper writeup, the isolation of new methods and ideas, and citing relevant previous work of others." Without that writeup step, the 25 Fields Medallists argue, it becomes unclear whose ideas a proof rests on, which is where the attribution and plagiarism questions start.

What does "severe misalignment" mean in the declaration's title?

The misalignment the declaration describes is between what AI companies optimise for and what mathematics is for. Producing true statements faster is the AI target; conceptual understanding, insight and the training of new mathematicians are the field's targets. The signatories warn that "mass production at faster and faster pace of 'true/false' statements could destroy fertile ground."

Terence Tao · 2026-09-11 · major

25 Fields Medallists sign a declaration — AI math benchmarks miss the point

Terence Tao published a declaration signed by 25 Fields Medallists criticizing AI companies for treating mathematics as a benchmark. Rushed AI solutions break attribution and skip the understanding research math exists for, the signatories argue.

Terence Tao speaking at an IPAM fireside chat in 2026 — Wikimedia Commons

25 Fields Medallists put their names to a declaration that AI's use of mathematics as a benchmark is harming the field.

Quick facts

Signatories	25 initial, all Fields Medallists
Published	11 September 2026
Posted by	Terence Tao, on his blog "What's new"
Full text	mathandai.org
Main concerns	Attribution, plagiarism, lost transmission chain

What is it?

A declaration signed by 25 Fields Medallists says the push by AI companies to solve open mathematics problems as a benchmark is damaging the discipline. Terence Tao published it on his blog "What's new" on 11 September 2026 and points to mathandai.org for the full text. All 25 are initial signatories; the post names no single lead author.

How does it work?

The argument turns on a distinction the declaration draws: "Solving problems is only a tool and proxy for achieving the primary goal of conceptual understanding and insight," it says, and "forgetting this in the world of AI may turn the tool against the primary goal." From there it describes how results now arrive — "announced in a rush, leaving no time for a proper writeup, the isolation of new methods and ideas, and citing relevant previous work of others."

Why does it matter?

Mathematics has become a public scoreboard for AI labs, and this is a collective response from 25 of the field's most decorated researchers rather than one person's blog post. Two concrete harms are named: attribution and plagiarism questions when a proof appears with no writeup, and the loss of what the signatories call the human transmission chain between mathematicians. Tao's post links a recent Economist article on OpenAI's methods as the immediate example.

Who is it for?

mathematicians and AI researchers working on formal reasoning

Frequently asked questions

Who signed the AI and mathematics declaration?: The declaration carries 25 initial signatories, all of them Fields Medallists. Terence Tao, himself a Fields Medallist, posted it on his blog "What's new" on 11 September 2026 and hosts the full text at mathandai.org. Tao's post identifies him as a signatory but does not name any single author of the declaration.
Does the declaration name any AI company?: Terence Tao's post points readers to a recent Economist article describing OpenAI's methods, and OpenAI is the company named there. The declaration's argument is aimed at the wider industry practice of using open mathematical problems as a capability benchmark, not at one lab's individual products.
Why do the signatories raise plagiarism and attribution?: The declaration says AI results are "announced in a rush, leaving no time for a proper writeup, the isolation of new methods and ideas, and citing relevant previous work of others." Without that writeup step, the 25 Fields Medallists argue, it becomes unclear whose ideas a proof rests on, which is where the attribution and plagiarism questions start.
What does "severe misalignment" mean in the declaration's title?: The misalignment the declaration describes is between what AI companies optimise for and what mathematics is for. Producing true statements faster is the AI target; conceptual understanding, insight and the training of new mathematicians are the field's targets. The signatories warn that "mass production at faster and faster pace of 'true/false' statements could destroy fertile ground."