Probabilist Ivan Corwin proposes a value-based approach to mathematics under AI, framing mathematical value across society, students, community, and individuals, and calling on the community to articulate that value to funders

The author proposes that mathematicians produce value in four loci: society (mathematical ideas and methods drive scientific and societal progress, as with the central limit theorem, extreme value distributions, Brownian motion, random matrices, and spin-glasses), students (rigorous mathematical training builds capacities such as working in abstract or logical systems, grappling with complex systems, working in open-ended areas, communicating complex ideas, and learning from repeated failure), community (mathematical communities prize understanding over results, are supportive and collaborative, value students, and tend to avoid hegemonies), and individuals (mathematicians derive considerable personal joy and fulfillment from their own process and work). Relative to discussions that measure mathematical value mainly by cutting-edge research output, the author extends the value framework to societal influence, teaching and training, community norms, and personal experience, and argues that the community should use this framing to articulate its value to funders and society. This is an opinion framework grounded in the author's experience and observations as a probabilist, illustrated with examples from probability theory, random matrices, and spin-glasses; it is an argumentative essay rather than an empirical study.

The author argues that improper use of AI 'mortgages' future mathematical development: students and researchers may not experience the growth that comes from failing and struggling with a problem, may become risk-averse, and may engage more with AI than with each other; the text notes an observed burst in arXiv postings and problems being solved by AI. The author moves the risk discussion beyond 'tool replacement' to the erosion of the learning process and community interaction, and argues that if AI is used to indiscriminately solve open problems without mathematicians being deeply involved, it may not produce much societal value from the process or even the product. The text cites the observed burst in arXiv postings and AI problem-solving as observational evidence, without specific statistics or controlled studies; this is the author's judgment and observation.

The author issues a call to action: the mathematical community should understand, articulate, and communicate the value of mathematics; ground discussions and decisions about mentoring and resource/reward allocation in long-term value; include younger members such as undergraduates and graduate students in these discussions; create new venues (e.g., journals and conferences) to articulate mathematical value to the broader public; and recenter training and teaching, seeing teaching as an opportunity to bring great value to students rather than as a matter of service to the university. Relative to leaving value discussions to senior professors and stewards, the author explicitly calls for including younger members and names efforts such as Essential Number Theory, the forthcoming Essential Analysis, Mathematical Discourse, Galileo, and Quanta, as well as research institutes (like SLMath, for which the author co-chairs the scientific advisory committee) as playing a key role in allocating resources in support of the value of mathematics. This is an initiative proposed from the author's personal experience and reflection, with no empirical data on implementation effects.

The author uses his own writing process to illustrate the value of 'process': he could have asked AI to write or help write and edit this blog, but then he would not have taken the time to formulate and solidify his thoughts, to share, discuss, reformulate, and rewrite, to think through arguments he ultimately abandoned, or to see this as a call to action for himself. The author makes the criterion for how much AI should be used concrete: AI should be used insofar as it enhances the value derived from both the process and the product. This is a first-person account of the author's own writing experience, constituting personal experiential evidence.

This article is aimed at the mathematical community, the public interested in mathematics, and those who control funding for mathematics, and it applies to discussions about the value of mathematics, the positioning of teaching, and modes of resource allocation. The four-loci value framework can serve as a starting point for discussion within the mathematical community and in collaboration with other disciplines and historians, and can be used to explain the long-term value of mathematics to funders. The author also recommends including undergraduates and graduate students in these discussions and suggests that research institutes support the communication of mathematical value in resource allocation.

This is an opinion piece whose judgments rest on the author's experience and observations; the burst in arXiv postings and AI problem-solving mentioned in the text are not accompanied by specific data. How the four-loci value framework will be received by the mathematical community and what effects the specific proposals in the call to action (such as new journals, institutional resource allocation, and including younger members in discussions) will have remain to be seen. In addition, the text read here is of incomplete scope; if the original contains figures, data, or further discussion, this summary does not cover them.