OpenAI reports sharp nodal-length bounds on smooth surfaces

For each fixed smooth closed connected Riemannian surface, the nodal set of every nonzero real Laplace eigenfunction of eigenvalue λ>0 has length at most C√λ, with C depending on the surface.

Sources for this record

What was already known

The known lower bound was already of order √λ. For smooth surfaces, earlier upper bounds included Cλ^(3/4) and then Cλ^(3/4−β) for universal β>0; the sharp upper bound remained Yau’s conjecture.

AI involvement AI-led

OpenAI credits itself as manuscript author and states that its internal frontier model produced the results in this release. Individual human input is not detailed enough to establish task-only autonomy.

OpenAI is named as author on the manuscript, and its release attributes the results to an internal frontier model. The individual human input is not detailed enough to establish autonomy.

Importance Major

Proves the missing upper half of Yau’s nodal-length conjecture for smooth closed surfaces.

Limitations

The result is two-dimensional and assumes a fixed smooth closed connected surface. The known lower bound is a separate earlier result.

Technical verification Formal (Lean 4)

OpenAI reports a Lean proof matching the selected claim.

Formal checking

The pinned Lean note covers the upper bound for every nonzero real eigenfunction on each fixed smooth closed connected surface.

Checker
OpenAI-reported Lean 4 / Comparator
Scope
For each fixed smooth closed connected Riemannian surface, the nodal set of every nonzero real Laplace eigenfunction of eigenvalue λ>0 has length at most C√λ, with C depending on the surface.
Basis
Reported by the linked checker
Evidence source

Checking a formal statement does not establish that it matches the original problem.

Human understanding Few

Few is the default when evidence for a higher level is insufficient. No named person is documented as explaining the central argument for this assessed contribution.

Correction and publication history

2026-10-18 · publication

Primary manuscript, attribution, prior-work, and scoped checking review for non-notifying local inclusion.

Read this published revision