Derek Thompson interviews mathematician Steven Strogatz about the grief surrounding AI-generated mathematical proofs. The concern is not just losing status or employment: it is losing a place in a conversation that stretches across generations.
Thompson separates three things that can come apart when language models produce discoveries:
- Getting an answer: a proof can settle a problem without giving people an accessible explanation of how to think about it.
- Understanding the answer: human insight helps turn one result into new questions and connections.
- Sharing the search: teaching, disagreement, and collaboration are part of the value of science, not merely overhead on the way to a result.
The essay leaves room for the upside: more people may gain access to mathematical discovery, and useful systems need not wait until humans fully understand their workings. Its argument is about what scientific work is for, not an independent audit of the mathematical breakthroughs it describes.
What the thread adds
The 248-comment thread on Hacker News adds a material objection to the essay’s hopeful ending, along with counterexamples to the idea that human mathematical work has lost its value:
- atleastoptimal — “How is one to feel safe in the progress of the future when the only hints of the wonderful post-scarcity utopia we can quantify so far is the removal of all human leverage via their labor?”
- vector_spaces — “But many more mathematicians have been toiling away in precarity and obscurity and without tenure for years and years. And of course many mathematicians went into private sector roles as software engineers, actuaries, analysts, ML engineers and similar, and have been displaced there, too.”
- bunderbunder — “Every so often I ask one to suggest a good binomial confidence interval for a particular application, and it still tends to just suggest one of the subpar options that everyone learns in a first year stats course.”
- omnicognate — “I happen to be (re)reading Strogatz’ book “Nonlinear Dynamics and Chaos” right now. It’s beautiful thing, and nothing in the “age of AI” threatens to equal it or in any way diminish my enjoyment of it.”
Handles are pseudonymous; these are commenters’ claims and experiences, not independently verified findings. HN publishes no per-comment scores, and the fetched ordering is HN’s own ranking. This is a slice of the thread, not a consensus.