'Breathtaking,' 'Devastating': Mathematics Reels After New OpenAI Release

Commenters are reacting to an OpenAI release that reportedly solved a large set of mathematics problems, with many voices focusing on who gets credit and control. sseagull argues the worrying precedent is that OpenAI, not academics, did the solving and that large companies could corner scientific advancement for profit, likening it to Amazon competing with the community that created ideas. novaRom calls the model “parasitic” for digesting human chat logs and presenting results as its own. perching_aix doubts the results without formal Lean proofs and warns closed-loop verification remains essential. marojejian highlights that humans solving such problems would merit top prizes and pxx reminds readers of the Fields Medal age rule, underscoring concerns about attribution and recognition.

Other commenters debate how society should respond. variadix and alphabeta3r56 demand government intervention or public-compute programs to prevent monopolization, while weard_beard urges collective protest and unionizing to pressure model owners. shadowgovt offers a more nuanced view, saying replacement can be acceptable if dignity remains and proposing a shift in mathematics toward human-shaped accessibility, a point credited to Grant Sanderson’s recent commentary. Opinions divide between those calling for regulation and resistance to protect researchers and those who see potential for positive reshaping of the field if incentives and dignity are preserved.