The first time I realized how much I could do by pasting scripts into ChatGPT, it felt like cheating. I could put the script into a chat window and get a working correction back. As a programmer, I was quite happy to spend less time dealing with a bad script.
So I kept using it.
Only after a few more useful experiences did I start to understand the scale of what I was looking at. I had welcomed help with one tedious part of programming. There was no reason its capabilities would stop at the parts I wanted to hand over.
That was a shock. This was reaching into my whole understanding of programming, and into the career I had built around it.
Reading mathematicians discuss AI, I think I recognize something in their wariness. This is an analogy from my own experience, rather than an explanation of what every mathematician feels. But I find it difficult to dismiss a reaction I have felt myself.
The help arrives before the larger question
The first decision was easy because it was small. Did I want help with this script? Yes. Did the answer work? I could check it.
The larger question arrived through repeated use. Each useful result gave me another reason to come back, and another reason to reconsider what I thought required my expertise.
I still use coding agents. I replaced ESLint with Biome because Claude Code or Codex could finish a patch before the linter finished checking it. By then, I was making the rest of my workflow keep up with the agent.
There is relief in that. There is also something unsettling about finding that a tool can do more of the work through which you learned to think. Both reactions can be honest at the same time.
Mathematics gives the feeling a different shape
On October 6, OpenAI released mathematical results produced by an internal model. Its repository includes preprints and formalized proofs. I have not independently checked the mathematics.
Jeremy Avigad’s guest essay on Terence Tao’s blog describes a profession reconsidering its daily work as results that once supported a respectable publication become easier to generate with AI. He also describes mathematics more broadly as a culture of rigorous reasoning and communication.
That distinction helps me understand the unease. A result can be valuable while changing the activity through which people developed their skill, found collaborators, or understood their place in a field.
In his October 6 thread, Tao describes what traditionally follows a breakthrough: talks, workshops, questions, new connections, and proofs becoming easier to understand and teach. Producing the answer starts a much larger exchange.
Programming has its own version of that exchange. I can read an implementation, ask why it was designed that way, and learn something I can use elsewhere. A finished patch is useful. Understanding the choices behind it is useful in a different way.
The comparison has limits. Software tests and mathematical proofs do different jobs. I am drawing a connection between people’s relationships with their work, rather than claiming the fields will change identically.
Useful and unsettling can coexist
I do not think someone has to stop using AI to take that change seriously. I also do not think finding the tool useful makes every reservation disappear.
Use it, and you may feel that you are participating in a change that unsettles you. Avoid it, and you give up useful help while the field may continue changing around you. That is the damned-if-you-do, damned-if-you-don’t feeling I recognize.
It is difficult to turn that into a clean moral choice. Someone who uses an agent may be trying to finish necessary work, learn something unfamiliar, or reclaim time. Someone who hesitates may care deeply about understanding and teaching. Those motives do not have to be in conflict, even when the available choices feel that way.
The original appeal remains real for me. I prefer getting the script fixed. I can welcome that relief and still feel the shock of realizing that the same capability reaches into work I care about much more.
That is why I want to approach mathematicians’ concerns with empathy. I would rather ask what they want to preserve and what the tools could help them explore than assume they simply have not tried a good enough model yet.
What happens after the answer?
Tao’s Math 2.0 proposal gives explanation, community building, and new directions more weight alongside raw problem solving. He sees AI contributing positively to those activities too.
OpenAI’s October 15 update also describes support for a Simons Foundation initiative to help mathematicians understand and build on AI-produced results. That is a concrete response to the question of what happens after a result appears.
In my own work, I can ask an agent to compare approaches, explain an unfamiliar design, or help test a prediction. I can also accept a patch and move on. Neither choice tells the whole story of what AI means for programming, but the way I arrange the work affects what I learn from it.
I do not know whether mathematics will follow programming along the same path. Repeated usefulness changed how I worked before I understood how much it might change the work itself.
I still want the help. I understand why that does not make the adjustment easy.
Was this useful?
Thanks for the feedback.