Claim 11 explained: the bare exponent d over (d plus 1)

Michael Darius Eastwood · Independent AI alignment researcher

Published 3 July 2026

What the claim says

The claim in this row is a piece of housekeeping the programme wants readers to see clearly. The exponent d over (d plus 1), sometimes called the "geometric speed limit" in the programme's own writing, is not being claimed as original. It has a long and honourable history in the theoretical biology and scaling-law literature. The programme uses it as an ingredient because it works: thirteen out of thirteen metabolic exponents fall inside the family, and the physics error against measurement is under 0.2%. The purpose of listing it as a spine claim rather than folding it silently into other rows is to name the prior art explicitly and to keep the exponent's originality separate from the unification argument in Claim 4.

The evidence

Papers III and VII, Origin and Foundational, all rely on d over (d plus 1) as a fitted family. The paper trail behind the exponent is longer than the programme's own. West, Brown & Enquist derived it in a landmark 1997 paper for metabolic scaling. Banavar and coworkers extended the family in 1999 and again in 2010. Demetrius developed the entropic-selection version in 2003, 2006 and 2010. He and Chen offered a network-flow derivation in 2003. Bettencourt derived a related exponent in the city-scaling literature in 2013. Maino and coworkers added a further variant in 2014. Zhao's 2022 preprint on arXiv (2205.14939) is a relevant solo-author contribution that is unpublished and has no DOI; the T5 verification record lists it in a footnote status rather than alongside the six published groups.

The honest caveat

The honest position on Claim 11 is that it exists to prevent a mistake. Any reader who sees the programme use d over (d plus 1) in the derivation of the Cauchy unification (Claim 4) and mistakes the exponent for the programme's own invention will get the priority story wrong. What the programme owns in this territory is the argument that the independent derivations named above reduce to a shared Cauchy structure with different boundary conditions. What it does not own is the exponent itself. Framing the exponent as a spine row in its own right, with a Prior-art synthesis status and named citations, makes that boundary auditable rather than easy to overlook.

What would kill it

The falsification contract for this row is the Cauchy unification contract, because that is where the programme's original claim actually lives. An independent classifier selects at least six domains with established scaling exponents from independent labs, extracts the effective dimension d implied by each exponent, and tests whether every exponent satisfies the Cauchy multiplicative form under that d. Confirmation requires at least five of six domains consistent at R-squared greater than 0.9. Three to four consistent is weakly supportive. Fewer than three refutes the unification and reduces d over (d plus 1) to what it has always been in the wider literature: a stable regularity of nature whose deeper cause is still open.