When Po-Shen Loh began coaching the United States Math Olympiad team in 2013, artificial intelligence was hardly on his mind. The professor at Carnegie Mellon University has always focused on human education, but recent developments in machine learning have inspired him to wonder about the future of math-capable AI.

In 2022 research startup OpenAI released their ChatGPT application, which could answer math and text prompts. Almost overnight, math teachers and professors worried about how this might harm education with either correct or incorrect solutions. But Loh remained skeptical that an AI algorithm could perform math at the highest levels. Until, in July 2024, an algorithm designed by Google Deepmind solved Olympiad problems well enough to achieve a silver medal. Loh's skepticism dissolved after the Olympiad advance, but he stresses that humans maintain a sort of creative edge over computers — particularly when it comes to recognizing the "beauty" of a math problem to solve.

As algorithms have matured, mathematicians have embraced computation as means to accelerate math discoveries. Researchers integrate AI with "proof checking" tools, like the software Lean, to avoid errors more quickly. According to Loh, this makes it easier for large teams to collaborate and for peers to check each other's discoveries for accuracy. According to a study in the journal Nature, mathematicians expect that AI will guide human intuition, and help to prove unsolved conjectures.

While Loh does not use AI directly in his mathematical research, it has sparked an urgency in him to teach students better and faster. Now, he hopes that humans will be able to keep up with AI.