Mathematicians Are Panicking About AI. They’re Terrified of the Wrong Thing.

You’ve seen the headlines. AI is coming for creative writing. AI is coming for coding. AI is coming for legal work. But there’s a quieter panic happening in one of the most elite, self-contained corners of academia: pure mathematics.

And the panic reveals something mathematicians don’t want to admit.

A recent essay titled “Mathematicians Need to Act” lays out the fear. AI tools are getting better at mathematical discovery. They’re proving theorems, finding patterns, doing the thing that mathematicians have claimed as their sacred territory for centuries. The essay’s argument? Mathematicians should resist. They should slow down AI’s encroachment on their field.

But here’s what happens when you read between the lines. The author’s justification for why slowing AI doesn’t matter — for why it’s fine if progress stalls — is that pure mathematics doesn’t actually produce tangible benefits for society. The only real output, the argument goes, is intellectual satisfaction for the mathematicians themselves.

If the only value of your field is the warm feeling it gives the people doing it, you don’t have a discipline. You have a hobby with a grant budget.

Think about that for a second. This isn’t some anti-intellectual outsider attacking math. This is a mathematician, from inside the field, essentially admitting that the work doesn’t need to justify itself to anyone else. The pleasure of the pursuit is enough.

Now imagine walking into a grant review meeting with that pitch.

“We’d like $500,000 to explore the topology of higher-dimensional manifolds.” “What’s the societal benefit?” “Well, it makes us feel alive.” “…Next applicant.”

The rest of science has to answer the funding question every single day. Biologists justify their work with medical applications. Physicists point to materials, energy, national competitiveness. Computer scientists have an easier time — their work literally builds the tools everyone else uses. Even philosophers, the most abstract of the bunch, at least gesture toward ethics, policy, human meaning.

But pure mathematics has built a culture where asking “what’s this for?” is considered gauche. The question itself is treated as a sign you don’t get it. Beauty is its own justification. Truth is its own reward.

That stance worked fine when AI couldn’t touch the field. It was a charming eccentricity. “Oh, those mathematicians, doing their pure research, isn’t it lovely that someone still values knowledge for its own sake.” The broader scientific community could afford to indulge it, the way a wealthy family tolerates an uncle who paints abstract canvases in the attic.

AI changes the equation. Not because AI can suddenly do everything mathematicians do — it can’t, not yet, and maybe not for a while. But because AI forces the question: if a machine can find the proof, and the proof has no external value, then what exactly are we funding?

The threat isn’t that AI makes mathematicians obsolete. The threat is that AI makes the question of mathematicians’ relevance impossible to dodge.

And here’s the twist nobody’s talking about. The mathematicians complaining about AI are the same ones who spent decades insisting their work needed no justification. They built a wall around the field — no practical applications required, no accountability to the public, no translation of results into anything usable. Pure thought for pure thought’s sake. A beautiful, hermetic garden.

Now someone built a machine that can tend that garden faster. And suddenly the gardeners are upset.

But they can’t have it both ways. You can’t say “our work has no practical value” and then say “society must protect our work from being automated.” On what grounds? If the output is purely internal satisfaction, then AI producing more theorems faster just means more internal satisfaction — for the mathematicians who read them. By the field’s own logic, AI is a net positive.

The real crisis isn’t technological. It’s existential. Mathematicians are being forced, for the first time in generations, to articulate why their work matters to someone other than themselves. And many of them are discovering they don’t have an answer.

Some do, of course. Number theory underpins cryptography. Topology informs data analysis. Combinatorics drives algorithm design. The connections are real and deep. But the field’s culture has spent so long dismissing those connections as incidental — as vulgar commercialization of sacred art — that it’s lost the muscle to make the case convincingly.

You can’t spend thirty years telling people your work has nothing to do with their lives and then be surprised when they stop paying for it.

So here’s what mathematicians actually need to do. Not “act” against AI. That’s protectionism, and protectionism in intellectual fields is always a slow death. What they need to do is the thing they’ve refused to do for a century: walk out of the garden, into the messy world, and explain — in language non-mathematicians can feel — why this work matters.

Not beauty. Not purity. Not the romance of the pursuit. Those are answers for the converted. The world needs to hear what mathematics does. What it enables. What breaks if it stops.

Because here’s the uncomfortable truth at the bottom of all of this. Every field that survives long enough eventually has to answer the funding question. The ones that answer it well thrive. The ones that refuse, that retreat into purity and pride, end up as footnotes — beautiful, brilliant, and forgotten.

AI didn’t create this crisis for mathematics. It just held up the mirror.

The question was never whether AI can do math. The question was always whether math could justify itself. And for too long, the answer has been silence.