What was already known
Generic-kinematics vanishing statements and the earlier single-minus gluon result did not supply this gravitational construction. The model was given the gluon paper as context. The later root-cone/cut-test reformulation already in the catalog is a different, subsequent contribution.
AI involvement
Lupsasca reports that GPT-5.2 Pro was asked to extend the gluon result to gravity and supplied the solution using the directed matrix-tree theorem plus a preliminary paper draft. The primary paper also credits an internal OpenAI model. Humans supplied the gluon context and continued checking and revision; task-only autonomy is not established.
Importance
Supplies explicit amplitudes and recursion/symmetry relations in a specified non-generic gravitational scattering regime.
Limitations
This concerns tree-level amplitudes on restricted kinematics, not generic physical 3+1-dimensional scattering, quantum-loop corrections or experimental confirmation of quantum gravity. Ward reconstruction uses chamber/analyticity assumptions. Later root-cone results retain separate credit.
Technical verification
Alexandru Lupsasca’s research account reports analytical verification of the final formulas and checks against known physical limits. It does not provide a named, complete proof-checking record for every distributional and Ward-identity step; the scoped result is rated Partial.
Human understanding
The paper and research account explain the physical setting and name the matrix-tree technique, but no sufficiently specific named human explanation of the central distributional tree construction, separate from the model-produced draft, was established. Few is the documentation default.
Correction and publication history
Added the sourced contribution with its stated hypotheses, AI role, prior work and checking limits.
Read this published revision