The formulation of a problem is often more essential than its solution, which may be merely a matter of mathematical or experimental skill. To raise new questions, new possibilities, to regard old questions from a new angle, requires creative imagination and marks real advance in science.
from The Evolution of Physics by Albert Einstein
On September 11, Terence Tao published a joint declaration titled “A Severe Misalignment of AI in Mathematics” on his blog, What’s New?. The manifesto was initially co-signed by 25 Fields Medalists, including Peter Scholze and Yu Deng.
However, to understand the inevitable trajectory of this perceived crisis, these eminent scholars need only look at the software industry. Over the past few years, legions of programmers have been forced to undergo a radical professional transformation. They have evolved from being manual typists of syntax into high-level architects who formulate problems and delegate execution. Instead of grinding through boilerplate code line by line, modern software engineers now define parameters, prompt AI models with precise logic, and rigorously review the generated outputs. They recognized early on that human value no longer lies in the mechanical production of code, but in the formulation of the right questions. The tech industry did not perish in the face of generative AI; it simply shifted human intellect higher up the cognitive stack.
The mathematical community must now confront the exact same paradigm shift. Mathematicians must learn to transition from their traditional role as sheer problem solvers to a new identity as question formulators and result verifiers. A look back at history reveals that the absolute greatest mathematical giants — figures like Bernhard Riemann, David Hilbert, Pierre de Fermat, and Henri Poincaré — achieved immortality precisely because they were the architects of profound questions, conjectures, and entirely new frameworks. They charted unexplored territories and set the agenda for centuries. In stark contrast, many contemporary Fields Medalists, despite their undeniable brilliance, have built their legacies merely by solving the towering homework assignments left behind by these intellectual ancestors. If AI can now mass-produce the grueling labor of proofs and logical derivation, the human mathematician’s mandate must return to the genesis of ideas: identifying what is actually worth proving.
The ruthless Darwinian logic currently reshaping computer science will invariably sweep through the ivory towers of mathematics. Just as in programming, only the mathematical practitioners capable of proposing genuinely profound questions and navigating the true frontiers of the unknown will survive this technological transition. The vast majority of ordinary practitioners — those whose careers are built on grinding through incremental derivations, routine calculations, and pedestrian lemmas — will inevitably be rendered obsolete by increasingly powerful AI systems. Terence Tao and his esteemed co-signatories must recognize and embrace this irreversible reality. The “misalignment” they lament is not a flaw in artificial intelligence, but rather their own reluctance to let go of an antiquated, artisanal era of mathematics. Clinging to the manual romance of proof-writing is a futile defense of the past; the future belongs entirely to those who know which questions to ask.