Quoting Terence Tao

The landscape of mathematical research is undergoing a fundamental shift as artificial intelligence systems increasingly tackle open problems that have long served as crucial drivers of innovation and discovery. Recent commentary from prominent mathematicians highlights a concerning trend: the finite pool of genuinely fruitful open problems may be diminishing faster than new ones can be generated, potentially reshaping how mathematical research is conducted and valued in coming decades.

The acceleration of problem-solving through AI systems has created an unprecedented situation where researchers face increasing pressure to secure open problems before computational approaches render them solved. This phenomenon extends beyond actual solutions—mere rumors of someone working on a particular problem can now trigger waves of competitive research activity and resource allocation. The traditional scholarly model, which valued incremental progress and collaborative exploration over extended periods, faces disruption as computational mathematics advances at exponential rates.

This scarcity is emerging as a "non-renewable" resource issue, where the historical accumulation of open problems across centuries of mathematical inquiry is being consumed faster than fresh, equally compelling problems are being formulated. The pressure to work on problems before they become obsolete or solved represents a qualitative change in research dynamics.

• AI systems are accelerating the resolution of problems that previously occupied mathematicians for years or decades
• Research institutions must adapt funding models and career incentive structures to accommodate faster problem turnover
• The competitive pressure may incentivize quantity over quality in problem selection and research approaches
• Mathematical education may require restructuring to emphasize problem formulation over traditional problem-solving
• Emerging mathematical areas and pure theoretical work might receive less attention and resources
• The traditional publication and peer-review timeline may become misaligned with AI-driven research velocity

This shift carries profound implications for mathematics and scientific progress broadly. Open problems have historically motivated theoretical development, interdisciplinary collaboration, and fundamental breakthroughs. If the supply of genuinely significant open problems becomes constrained, the research community faces questions about sustainability, career viability for mathematicians, and whether mathematics itself risks exhausting its frontier problems—a scenario that seemed theoretical mere years ago but now warrants serious consideration.