OpenAI Pulls From Caltech Math Hackathon Amid AI Slop Criticism
Left 67%
Center 33%
Right 0%
2 left · 1 center · 0 right
What happened
On Thursday, OpenAI said it is no longer sponsoring a mathematics hackathon at the California Institute of Technology, after current and former Caltech mathematicians issued an open letter on Wednesday warning that AI-company-sponsored events and rushed AI claims on unsolved problems would harm the field. OpenAI research lead Dan Roberts said the company recognizes that rapid AI progress in mathematics is disruptive and wants to engage mathematicians on how to integrate the technology. The undergraduate-run event, also backed by Anthropic, was designed for student teams to use large language models on open research problems; the two companies had agreed to provide $2 million in AI credits. The withdrawal followed OpenAI’s public claim that its agents had produced a solution to the Navier-Stokes equations, a Millennium Prize problem, and public accusations from NYU mathematician Tristan Buckmaster about the company’s dealings with researchers. Event organizers said they did not intend to cause controversy and that the hackathon was meant to keep young people engaged with mathematics.
Sourcing check · Mostly on the record · 3 articles
2 of 3 articles cite a named source for the story. It was on the record from the first report. Laurance Fauconnet denied part of it on the record.
What these terms mean and how they're counted
- On the record · names a source 2 of 3
- On the record means the article cites a source by name for the story's main claim: a person, or an institution speaking officially, such as a company statement or a named agency's spokesperson. The opposite is an unnamed source. The count is how many of the counted articles cite at least one such source; the names are everyone named across them, each once however the articles spell it. Here: Dan Roberts, Tristan Buckmaster, Andreas Thom and current and former Caltech mathematicians
- Primary evidence · cites a document or data 2 of 3
- Primary evidence means the article cites a record you could check yourself: a court filing or ruling, a dataset or report, a recording or transcript, an official statement or press release, market data. A quote is not a document. The count is how many articles cite one. Here: open letter, statement
- Credits another outlet 0 of 3
- The article says the story comes from another outlet's reporting (“first reported by Reuters”). Those articles relay the story; they are not independent confirmation. Here: Each article presents its own reporting.
- Unnamed sources 1 of 3
- The main claim is attributed to people who are not identified: “people familiar with the matter”, “officials said”. The count is how many articles do this.
- On-record denial 1
- A named person or institution says the claim, or part of it, is not true. It is on the record — someone signed their name to it — but it is not confirmation, so it never counts toward On the record. Criticizing or dissenting from a decision is not a denial. The count is how many people or institutions deny it; each is listed with the first article that carried the denial. Here: Laurance Fauconnet (The Verge, Sep 12, 11:00 UTC)
- How it developed
- When the first report landed and whether it named anyone; when the first named source and the first document appeared. Here: First report Sep 10, 22:46 UTC by Business Insider, naming its source. First document in the first report, from Business Insider (open letter).
- Not counted 0
- Opinion, analysis and video pieces, and articles too short to assess. Every count above is out of the counted articles. An article republished under a new headline is one article.
This is about who is cited, not whether the story is true. Each source below says what it names. How we count
BLINDSPOT.
Only left-leaning outlets are covering this story
— the other side's media is silent.
Omitted — what each side leaves out
Unpacked
Right-leaning outlets had not covered this as of publication, leaving their readers unaware that OpenAI dropped sponsorship of a Caltech undergraduate math hackathon after an open letter from campus mathematicians. Left-leaning coverage reports the letter warned the event would dump unpaid verification work on researchers and serve as advertising for AI firms. Those outlets disagree among themselves on what matters: one centers the pullout and the student organizers' reply that they did not mean to cause a stir, while the other mentions the Caltech exit only while narrating a race between labs over a long-open fluid problem. A reader of the silent side is missing the sponsorship drop and that internal split.
Left-leaning coverage also splits on what to put first. Business Insider opens on the hackathon exit. The Verge leads with "OpenAI just wants to win" and treats the Caltech withdrawal as a late aside.
Who pays for the student tokens, and does the contest still run, now that OpenAI has withdrawn?
Bottom line
A reader of only right-leaning media would not know OpenAI left Caltech's math hackathon, a withdrawal Business Insider put in its headline after mathematicians criticized AI math slop.
The receipts — 5 facts one side's coverage lacks
- OpenAI withdrew its sponsorship of Caltech's undergraduate math hackathon
- Caltech mathematicians' open letter said the event is likely to have destructive impacts for the mathematical community
- Anthropic and OpenAI agreed to donate $2 million in AI credits to the event
- Student organizers wrote that they did not intend to cause such a stir
- First-round teams receive $20,000 in tokens to solve an open problem
The Left View
Business Insider, The Verge, and TechCrunch report that OpenAI ended its Caltech sponsorship after mathematicians attacked what they call slop mathematics: AI labs announcing solutions to famous problems and leaving verification, write-up, citation, and integration to uncompensated researchers. Coverage ties the pullout to a Caltech open letter, a letter from 25 Fields Medalists, and the June Leiden Declaration on responsible AI in math. The Verge, citing interviews with Tristan Buckmaster and Andreas Thom, frames OpenAI as racing Anthropic for Millennium Prize prestige, reports Buckmaster’s claim that he was offered compute and authorship if he dropped an Anthropic-affiliated collaborator, and notes OpenAI’s denial that his recent Codex prompts influenced its system. TechCrunch stresses that OpenAI’s Navier-Stokes proof remains unverified, that mathematicians fear industrial-scale scooping will push the field toward secrecy, and that similar credit and tool-provider conflicts may spread to other disciplines. Outlets present company statements alongside researchers’ accounts of incomplete attribution, rushed announcements, and conflict-of-interest concerns because the same firms supply research tools and compete with academic groups.
Our Take (balanced)
The Caltech sponsorship withdrawal is a substantive, checkable story: a named company publicly ended support for a named event after a dated open letter, and OpenAI’s Dan Roberts put the decision on the record. Also established are OpenAI’s Navier-Stokes announcement, Anthropic’s remaining sponsor role, the Fields Medalists’ letter, the Leiden Declaration, OpenAI researcher Sébastien Bubeck’s acknowledgment that the company offered Buckmaster resources or authorship help, and OpenAI’s later statement that Buckmaster’s Codex prompts from the prior two months could not have influenced its system. Still alleged and disputed are claims that chatbot conversations trained or steered the breakthrough, that the resource offer was a bribe or threat, and that user data generally fed the result; those rest on contested personal accounts and incomplete public documentation, not on released records. No right-leaning outlet has reported this. Watch next for the Clay Mathematics Institute’s review of the Navier-Stokes claim, any independent accounting of training-data questions, fuller on-record accounts from OpenAI and Anthropic of the disputed talks, and whether the Caltech event proceeds and how remaining credits are handled.
3 sources
More stories
The week's bottom lines, in your inbox
One email a week: the five stories that mattered and what they actually mean. Free.