The professor who answered OpenAI's proof dump with a music video
Updated:
On October 10, 2026, Joshua B. Plotkin — the Annenberg Professor of Natural Sciences at the University of Pennsylvania — posted a nearly five-minute music video with a two-line caption: "It saved some for you.
(One shot!)" Ethan Mollick, the Wharton professor who studies AI, shared it the next day with a one-sentence review: "A mathematician makes an Opus 5.5 music video in response to OpenAI's proofs. Pretty great." The video, per Mollick, was made with Claude Opus 5.5, Anthropic's flagship model released on September 22.
Plotkin's music video (about 4 min 53 s, 1080p)
That is the entire verifiable surface of the story: a professor, a video, two captions. But the event it responds to — and the fight it lands in the middle of — is the biggest controversy in mathematics right now. To understand why a music video counts as a position statement, you need the context.
What OpenAI did
Earlier in October, The Decoder reported, OpenAI published hundreds of AI-generated mathematical manuscripts — more than 700 files released at once — produced by an internal model. The company had already caused a stir by claiming that model solved more than 100 open math problems in a single month, including an approach to the Navier-Stokes Millennium Problem. OpenAI also established an advisory group, AGMAI, at the Institute for Advanced Study — though the group explicitly has no say over the pace of the company's internal research. (OpenAI's own announcement of the release is on its site; claims below follow The Decoder's reporting of it.)
The release landed on a field that had been arguing about this for months, and it detonated.
The backlash: "Mathematicians did not ask for this work to be done"
The Association for Human Mathematics (AHM) — chaired by Fields Medalist Terence Tao, who shared the statement as a guest post on his blog — accused OpenAI of violating core norms of scientific research and urged mathematicians to stop working with the company. The statement described releasing more than 700 files at once as a show of force rather than scholarship. Tao, separately, sketched a "Math 2.0" era on Mastodon in which solving problems should no longer be the discipline's main focus. Complexity theorist Scott Aaronson called the situation a "Mathocalypse" on his blog.
Two details in The Decoder's reporting explain the temperature. Many of the AI-generated papers are so dense and so oddly written that leading experts need AI assistance to decipher them — if they can at all. And MIT's Justin Solomon, discussing the model's Navier-Stokes counterexample on a podcast (as summarized here), said he and the host could not get through the second page. The dispute, at its core, is over whether a problem counts as solved before its proof has been independently worked through and made understandable to humans.
Which is precisely where Plotkin's video comes in.
The third response
Strip away the surface, and the mathematical world has produced three distinct responses to the proof flood. The first is institutional resistance: the AHM boycott, the appeal to norms, the insistence that understanding is the product. The second is grim acceptance: Aaronson's "Mathocalypse," Solomon's page-two surrender, the acknowledgment that the gate has opened whether or not anyone asked.
The third response is Plotkin's: a piece of art, built with an AI model from the rival lab, whose caption — "It saved some for you" — reads as a wry reply to the entire debate. Whether that line is a lyric, a joke about what machines leave for humans, or both, the gesture itself is the argument: when the machines handle the proofs, what's left for people is exactly this — meaning, performance, humor, taste. And it is made, pointedly, not by refusing AI but by using it.
Mollick's "Pretty great" places him on this side of the argument as well. He has spent years arguing that AI's significance is as a co-intelligence for human work rather than a replacement for it, and sharing a music video as his commentary on the proof wars is a consistent move: the interesting question is not whether machines can prove, but what humans do next.
The honest caveats
Three things this article cannot verify. The video's actual content — whether it is animated, performed, or generated — could not be reviewed here; only its length, caption, and attribution are on record. The claim that Opus 5.5 made it rests on Mollick's tweet; Plotkin's own post does not name the tool. And Mollick calls Plotkin "a mathematician" — his chair is in the natural sciences — a reminder that even the basic labels in this story come from social media rather than from any publication.
What is not in doubt is the timing and the structure: a proof dump, a boycott, and a music video, all within the same week. The first two were about what mathematics is for. The third quietly demonstrated one answer.