The Five Phases Described by Altman
During a talk at Stanford University before hundreds of students worried about their future employment, OpenAI CEO Sam Altman recounted, as the source reports, how the mathematics community is reacting to the entry of artificial intelligence into research. According to Altman, the path resembles mourning: first astonishment, then fear that the discipline is over and the idea of having to convince doctoral students to change course; around the third day comes anger toward OpenAI; finally, around the fifth, acceptance of AI as a now indispensable tool for research and proofs, with the hope of a different but better professional path. This is a reconstruction made by Altman and reported by the source, not a survey of mathematicians.
The CEO's thesis is that AI will not cause an employment disaster and that mathematics is the field where the transformation has occurred at the highest speed ever observed: he himself speaks of an unprecedented dynamic. The significance lies in the fact that it is an executive of the company that developed AI describing the adaptation of scholars as inevitable.
The Disputes: From the Conjecture to the Marathon
The CEO's optimism is not shared by much of the scientific community cited by the source. In September, OpenAI announced that one of its models had tackled a well-known mathematical problem linked to the Navier-Stokes equation, unsolved according to the source since nineteenth-century research. The announcement prompted a reaction from researchers at New York University and Anthropic, who disputed the use of their exchanges with the AI as data useful to the result. Other scholars complained that, even granting the solution, the derivation was not shown, leaving humans only verification.
The tension continued with a 'mathematical marathon' launched by Caltech students to tackle open problems with the aid of AI. More than 700 scholars signed an open letter criticizing the initiative for the risk of low-quality generated content; OpenAI withdrew its sponsorship and the organizers changed the rules, with the competition scheduled for November. In parallel, Professor Terence Tao of UCLA spoke of the end of the 'romantic' era of mathematical research, noting how many colleagues move from solving problems to verifying AI's answers, and commenting with a 'something is being lost' on OpenAI's latest announcement about more than 700 problems tackled.
What remains to be verified
The picture that emerges sets Altman's narrative, centered on adaptation and benefits, against the methodological and attribution objections raised by researchers. However, some points remain to be clarified: the source reports the disputes without indicating outcomes or independent verification of the announced solutions, and it is not known how many mathematicians actually share the five phases described or the criticisms. The November competition, after the rule changes, and the evolution of the relationship between human research and automated tools will be the elements to watch to understand whether Altman's prediction will be confirmed by events.
