The AI Revolution in Math Has Arrived (www.quantamagazine.org)

In July 2025, artificial intelligence models astounded the mathematical community by solving five out of six problems at the International Mathematical Olympiad, a notable achievement that created a buzz around AI's capabilities in mathematics. This milestone marked the beginning of a transformative era, as mathematicians began to leverage AI not only for puzzle-solving but also for groundbreaking research, realizing that these models could assist in formulating and proving new mathematical results much faster than traditional methods. Terence Tao, a prominent mathematician, emphasized that AI's utility was revolutionizing how math is conducted, allowing researchers to tackle thousands of problems simultaneously and implement statistical analyses, fundamentally altering the discipline. As 2026 approached, AI models such as AlphaEvolve demonstrated impressive acumen in optimizing complex problems by utilizing genetic algorithms and becoming adept conversation partners in discussions about mathematics. These advancements signal a cultural shift within the mathematical community, prompting some mathematicians to transition to tech firms and startups focused on AI for math. With the emergence of larger language models (LLMs) becoming critically engaging tools in mathematical exploration, researchers acknowledged that while AI would not replace mathematicians, embracing these powerful tools is essential for keeping pace with evolving methodologies. This shift presents both opportunities and challenges, as experts caution about the potential loss of deep mathematical understanding amid rising reliance on AI solutions.