In September, OpenAI announced that it had used autonomous agents to solve one of the $1 million Millennium Prize Problems posed by the Clay Mathematics Institute. If it holds up, this is a momentous occasion for math and for AI. This advance and other similar events have led some people to declare that math is forever changed and that mathematicians will be out of work because AI can do math better than humans.

Ravi Vakil, the Robert Grimmett Professor of Mathematics in the School of Humanities and Sciences at Stanford, advocates for nuance in the face of catastrophic predictions. He emphasizes the importance of humanity in math and is, overall, optimistic about how AI could be used in his field, so long as people apply it with care.

“We’re in this new world where things are changing. The people who are going to be born in it are going to be very comfortable with it. For those of us who are here for the transition, it’s going to be an exciting time, but also a disruptive and challenging one,” said Vakil. “The important thing is to remember why we do science and to keep doing it, and to protect the young people who are going to be the ones doing the science in the future.”

Stanford Report asked Vakil four questions about how AI is really affecting mathematics and what he thinks people should focus on in order to find the best path forward in this field.

Will AI ‘solve math’?

Saying “AI will solve math” doesn’t make sense. The nature of math is that there are always more questions to ask, and much of what we learn from math happens in the process of solving it, not simply from knowing the answers.

What has happened is that a bunch of major challenges that mathematicians have been fighting with for a long time are falling because of AI tools, and that is exciting. How we do our work is suddenly speeding up as well because of AI. A lot of things that were no fun to do, but were important to do, became easier. But some things that were fun to do are also becoming easier. So, it’s good and bad.

AI is accelerating mathematics at a rate that has not been seen for a very long time. Mathematicians can work much more efficiently; even those who don’t use AI can build on results that do use AI. Everything is potentially speeding up, and we need to figure out whole new ways of doing science.

What can humans bring to math that AI can’t?

Mathematics is a means of understanding the world, and it’s humans who want to understand the world. AI doesn’t want to know answers to things. The best results that have used AI are human guided – the tools are answering questions that humans pose. As things accelerate, we will continue to be responsible for that acceleration, and we need to learn how to use these tools wisely and effectively, and in nonharmful ways.

Mathematics has an extremely long history. Thousands of years. The subject has seen enormous changes over its history, but on human timescales, it looks like it doesn’t change at all. One thing which is different with this advance is the speed with which it’s happening. Even when computers came to mathematics, that was digested over the course of decades, whereas now we’re building infrastructure over the course of weeks. It will be up to us to decide how to digest these new tools incredibly quickly, and that is something we’re not used to doing, being such an old – and in some ways slow-moving – profession.

How do you think AI will change math, and how do you feel about it?

I have mixed feelings about AI’s effect on mathematics. On one hand, I’m a truth seeker, so the fact that we can now do science and mathematics much faster is a massive win. However, some of the things I enjoy about mathematics could become less common. I think about it like this: I have a lot of friends who loved driving stick shift. Automatic transmissions are, in many ways, better, but some of the joy of driving is taken away. It’s good that we still have cars with stick shift, and it’s the right option for when you need certain types of performance. But automatic transmissions are better for most purposes. The future is coming, and we should try to be comfortable with it.

Anyone who makes any confident predictions as to how mathematics is going to change should not be believed because there’s just no way of knowing. The uncertainty is really off the charts.

All the evidence, historically, is that it will be a very positive change because we can be much more effective at doing science and doing mathematics. That said, we need to protect younger mathematicians and people who are entering the field. These are the people who are potentially at risk if we are stupid about new technology and we don’t properly use it.

What advice do you have for young mathematicians who are worried about the future of math?

The worries that I see most often voiced are not the ones I think we should worry about.

The way in which some people do mathematics is going to change. That is clear and somewhat uncomfortable. The uncertainty is high, and human beings don’t work well with uncertainty. Being a tenured professor, I don’t have that uncertainty. But my students, of course, do. The most important thing in the next couple of years will be to build institutions that protect young talent because the next generation of scientists and mathematicians are coming up today. If we are at risk of losing them, then that’s the future of civilization that’s at risk.

I don’t worry about whether there’ll be jobs for scientists in the future. If history is any guide, whenever there’s new technology, there’s always a desire for more people who understand the subject, not fewer. Every time there’s been a technological revolution, the result is that we needed more people who could think quantitatively and mathematically and rationally.

I do worry about whether people will use this as a Trojan horse for anti-intellectualism and cutting the teaching of very fundamental forms of thinking. There are some people who argue that citizens no longer need to understand science. That citizens no longer need to know how to make an argument or how to read a book. If we give in to those ideas, the cost to society is going to be enormous. If we can fight that off, then the future is going to be bright.

On the whole, I feel like this is going to be the most transformative time in the history of mathematics because things are changing faster than ever before, and I’m very cautiously optimistic because history tells us that, at times like this, we should be optimistic. I’m also very worried that these changes put us at risk. Which way the future goes is up to us – not just the mathematicians and scientists, but also voters, administrations, and citizens – as we decide what the future of society is. During this time of change, we have to make some choices, and we need to really be careful to make wise, informed choices in a way that protects our basic values.