Po-Shen Loh · 2026-09-19 · notable
Po-Shen Loh — why we still need human mathematicians
Po-Shen Loh argues that every field wanting to stay human-led should adopt one axiom: humans should help humanity flourish. The Carnegie Mellon mathematician says AI oversight will create more skilled jobs than there are people to fill them.
Po-Shen Loh argues that AI oversight, not faster theorem proving, is what keeps human mathematicians necessary.
What is it?
A single axiom is what Po-Shen Loh asks mathematics to adopt: we (humans) should help humanity flourish. Writing as a guest on Terence Tao's blog, the Carnegie Mellon mathematician accepts that AI is getting very good at proofs and asks what is left for people. His answer is stewardship rather than speed.
How does it work?
The argument runs through what Loh calls control points. Frontier AI systems are, in his words, incomprehensible in a way ordinary software is not, so they cannot be supervised passively — every place AI touches something that matters, from banking to water systems, needs a person with current expertise watching it. Loh writes that the advance of AI will overwhelm us with so many control points to watch that there aren't enough people to control them all, and that those are jobs, highly skilled jobs. Staying qualified for one, he argues, means staying an active practitioner.
Why does it matter?
The shortage of qualified overseers will itself slow AI down, Loh concludes, which turns 'AI is taking our work' into a question about who is left to watch the machines. He sums up the limit with a driving analogy: going faster than you can run is fine, but not faster than you can steer. The essay is the second AI-and-mathematics guest post on Tao's blog in two days, after Grant Sanderson's on 18 September.
Who is it for?
mathematicians and AI researchers