The proper hat-guessing number of
Determine the exact proper hat-guessing number of the complete graph on seven vertices with one edge removed. The general bounds leave .
- Result
- Proved(see note)
- Status
- Candidate (review pending)
- AI contribution
- AI co-developed
- Method
- Construction
- Field
- Graph theory; hat-guessing games; Steiner systems; permutation groups
- Posed by
- Adriaensen et al.
- Year posed
- 2026
- Years open
- 0y
- Solved
- 2026-09-03
- Model
- GPT-5.6 Pro
- Vendor
- OpenAI
- Collaborators
- —
- Verification
- Unreviewed
- Publication
- Announced
- Significance
- 7 / 100
- Disclosed cost
- —
- Wikipedia
- No dedicated article
What was actually shown
We prove . One lower-bound proof uses two explicit block-disjoint Witt designs. A second uses orbit maps from an explicitly regenerated sharply five-transitive twelve-point permutation group, combining one set-symmetric and one order-sensitive rule. Both satisfy a general coordinate-line twin-completion criterion, and Hall's theorem completes the clique strategy. The release also proves a disjoint completion-design theorem, an even- obstruction scoped to set-symmetric line-permutation twins in this sufficient framework, and a prime-admissibility theorem for the design parameters. It does not solve the general problem or .
What the AI did
OpenAI GPT-5.6 Pro contributed substantively to literature search, target selection, construction search, proof development, code generation, exact verification, adversarial critique, and manuscript preparation. The Witt-design and orbit-map proof architecture was developed in a human-directed model-assisted research process. Matthew Protti selected and framed the target, directed the research programme, evaluated candidate arguments, commissioned independent adversarial review, required exact checks, approved the public scope, and accepts responsibility.
Verification
Unreviewed on this site's ladder, but better evidenced than either sibling. An independent adversarial review rebuilt both 132-block S(5,6,12) completion designs, all 792 pentad completions, design disjointness, the 95,040-element group, sharp five-transitivity, all 495 set lines, all 59,400 ordered lines and both Hall-degree censuses, returning ACCEPT_K7E_THEOREM with no mathematical repairs required; STATUS.json pins the review package by SHA-256. That is a real check by someone other than the author, but it is an adversarial review of an artefact rather than a named expert endorsing the theorem or a proof assistant checking it, and this site has not rebuilt it. Conventional peer review is not claimed.
Sources
- PaperFour-page K7-e preprint source
- CodeReviewed K7-e public research disclosure v0.1 (GitHub)Verifier, certificates and public disclosure
- AnnouncementPublic draft PR #2 and release record
- Problem recordHat guessing with proper colorings (Adriaensen et al.)
- OtherIndependent adversarial review acceptance
Submitted by Matthew Protti on