Submit an entry
Know of a problem proved or disproved with an AI model in the loop that isn't tracked here? Propose it. A reviewer checks the source before it goes live, and the published entry credits you by pseudonym.
Sign in to submit an entry
Submissions are credited to their contributor and reviewed before they appear, so they need an account. It takes one click, and you are returned straight to the form.
You post under a random pseudonym. Nothing from the account you sign in with is shown on the site.
Not sure it qualifies? The methodology has the full criteria, and reading it first is worth more than a submission that gets turned down.
What belongs here: a precisely stated open question whose answer is now a proved or disproved theorem, with AI in the loop - any field of mathematics, theoretical computer science included. Not in scope: formalizations of known human results, empirical attacks, or improved heuristics. Every entry needs a real, checkable source - an announcement, an arXiv preprint or an article, not a summary of a summary. If a result is contested, unreviewed or partial, say so in the status and verification fields rather than leaving it out. Submissions are reviewed before they appear. You can submit up to 25 entries per rolling 24 hours. Full criteria in the methodology.
Fills the source, date and authors, and offers the title. You review every field before sending; nothing is submitted for you.
The result
What was solved, and where it is written down. Only you can supply this.
Math renders: $inline$ works here and shows properly on the site. In browser tabs, feeds and search the math degrades to plain text automatically.
Compact label for highlights, related-entry rows and chart axes. Math works: $inline$ renders everywhere except chart axes, where it degrades to plain text.
Nothing to preview yet. Plain text is fine; use $...$ for inline math and $$...$$ for display math.
Select the LaTeX preview tab to see a read-only rendering.
The problem as it was posed, in plain language. Not the paper's abstract, and not what was proved - that goes under "What was actually shown". Math works: $inline$ or $$display$$.
YYYY, YYYY-MM or YYYY-MM-DD. For a range, the completion date.
Where the RESULT lives: the paper, repository or announcement that proves or disproves it. Not where the problem was posed - that goes under Posed by.
What the source URL is, in a few words: "arXiv preprint", "GitHub repository with Lean proof", "OpenAI announcement".
The AI's part
Which model, what it actually contributed, and how checked the mathematics is.
How much of the mathematics the model contributed, going by the authors' own disclosure - pick the lower tier when it is vague. Writing or proofreading alone is out of scope entirely.
Nothing to preview yet. Plain text is fine; use $...$ for inline math and $$...$$ for display math.
Select the LaTeX preview tab to see a read-only rendering.
Math works here too.
How checked the mathematics is. Changing this requires updating the note below in the same edit, so the reason is on record.
Where the claim lives: a bare announcement, a preprint, or a peer-reviewed venue. Changing this also requires updating the verification note.
Nothing to preview yet. Plain text is fine; use $...$ for inline math and $$...$$ for display math.
Select the LaTeX preview tab to see a read-only rendering.
The prose explaining how strongly this result is checked.
Context
Helpful but not essential - a reviewer can fill or correct any of it.
The bucket the field filter groups by.
Free-text subfield shown on the card, e.g. Additive combinatorics.
What happened to the problem. Changing this requires updating the verification note in the same edit.
Who first asked the question, and where if it is known: "Erdős (1957)", "Fitzpatrick and Nowakowski, Problem 43".
Four-digit year, or blank if unknown.
Comma-separated names.
The decisive step: an explicit object, a finite certificate or case analysis, or a conceptual proof. Classify by what the result hinged on.
Nothing to preview yet. Plain text is fine; use $...$ for inline math and $$...$$ for display math.
Select the LaTeX preview tab to see a read-only rendering.
What the result does and does not settle: the caveat, the scope, what stays open. Math works: $inline$ or $$display$$.
Beyond the primary source: Lean repositories, independent proofs of the same theorem, verifiers, community records.
Nothing to preview yet. Plain text is fine; use $...$ for inline math and $$...$$ for display math.
Select the LaTeX preview tab to see a read-only rendering.
For the reviewer, not the entry - a duplicate you noticed, a related result, anything that doesn't fit above. Not published.
Rarely neededSkip unless one genuinely applies. Most entries leave every one of these blank.
If this is a numbered Erdős problem. Blank otherwise.
Nothing to preview yet. Plain text is fine; use $...$ for inline math and $$...$$ for display math.
Select the LaTeX preview tab to see a read-only rendering.
Only when a documented issue exists with the claim (refuted lemma, misformalized statement). Renders as a visible flag.
Whole dollars, only when a source actually states what the result cost. Never estimate it from model pricing - leave blank if undisclosed.
What the figure covers and where it came from, e.g. an aggregate across several results. Hover-bubble text, so plain text only - no math.
Shown in a hover bubble, so plain text only - write π/2, not $\pi/2$. Math renders in the statement and the prose notes, not here.
Looked-up count, or blank.