Pinned Tweet
Take a look at this weird integral. #Mathchallenge
If AI can solve Millennium Prize problems, I think my request is rather humble: I simply want Perron’s book on continued fractions translated from German into English in one shot, without errors and with no human intervention. Is this possible with current commercial AI models?
1
108
It has always amazed me how math slop drops are almost completely ignored by forums like Math Stack Exchange and MathOverflow. Attempts to discuss alleged AI proofs of the RH are usually ignored or even erased or downvoted
1
94
This is the most successful post about the topic I've seen so far: someone asked how to verify big AI math outputs and tagged it as a "soft question": mathoverflow.net/questions/5…
1
47
This looks like AI math slop, but it’s not! There’s actually a human way to solve it, although the solution is a bit lengthy. Give it a try!
Queryable data for StackExchange and MO is expected to be updated today. I get the impression that users have almost completely ignored the AI math release. Apart from mentions of the Flint Hills series and the quasi-RH, someone asked whether the results are unconventional #math
1
177
Heuristic: IA should make life easier, not harder. Things are getting surreal: two mathematicians, one whom is cited in one manuscript, were unable to make sense of proofs in areas related to their own fields.
OpenAI has released many math preprints, attracting considerable attention. As a mathematician who regularly uses various AI models including OpenAI's, I had a look at some of their preprints in my own field (combinatorics, like graph theory). My impression is that the quality of the written exposition varies considerably. Even before assessing their mathematical correctness, some are so poorly written that I cannot even tell what is being claimed. (My comments concern only the written preprints, not the Lean formalizations.)
For example, consider "165. Exact crossing numbers of complete and complete bipartite graphs." I find Lemma A.1 and its proof in p.11–p.12 of the following preprint ("The crossing number of complete graphs") particularly hard to follow. In my view, the exposition is not yet at a level suitable for peer review. If I were reviewing the paper, I would recommend major revisions to this part, at least.
github.com/openai/math/blob/…
Interestingly, the corresponding lemma in the other preprint on complete bipartite graphs appears more readable, although I have not examined that paper in detail.
github.com/openai/math/blob/…
How readable have you found OpenAI's preprints in your own area of mathematics? Have you noticed similar issues?
177











