Work

Inflation from the Running of the Species Scale: an SL(2,Z) Plateau along the String Coupling, a Tilt from α′³, and the Dimension at Horizon Exit Written into r

Year: 2026
Type: preprint
Abstract: Status update, 20 November 2026. This version adds a one-page addendum (24378141_addendum.pdf) after an independent referee check: verified errata, what has changed since publication, and links to the newest records. The original files are unchanged. Newest records: SERAPIS v3.4 status update (doi:10.5281/zenodo.24396665); The Quark Thread (doi:10.5281/zenodo.24399408); Weave Control and the Evidence Atlas (doi:10.5281/zenodo.24399447). v1.1 (19 Nov 2026, referee pass): citations completed (CMB-SPA: SPT-3G Collaboration, Camphuis et al., arXiv:2506.20707; species-scale normalisation: Castellano, Herráez, Ibáñez, arXiv:2310.07708); the claim that the string coupling is the only plateau direction across thirteen reductions is now stated as argued, with the heterotic entries below eight dimensions by analogy. No numbers changed. Companion papers: flavour doi:10.5281/zenodo.24377945; dimension ladder (forthcoming). Inflation from the running of the species scale (draft, 19 Nov 2026). We study a single-field inflationary potential built from the type IIB R⁴ coefficient E(τ,3/2). The inflaton is the string coupling. The potential is the squared running rate of the quantum-gravity cutoff Λ_QG = M_P E^(−1/6), normalised to its asymptotic rate 1/√8 (the species-gradient principle). It has the form of the Casas–Ibáñez one-loop tower potential, but its prefactor is not flat along the dilaton in 4D N = 1 supergravity, so here the principle is assumed, not derived. Given the principle: V = V∞(1/3 + (4/3)∇² ln Λ_QG) exactly; V → V∞[1 − (16ζ(2)/3ζ(3)) y⁻²] with no free shape parameter; V = 0 exactly at the self-dual point τ = e^(iπ/3), where inflation ends (confirmed by a two-field simulation of the end of inflation). A one-parameter tilt of the form produced by the BBHL α′³ correction places it in the observed band: exact Mukhanov–Sasaki gives n_s = 0.97065, r = 5.41×10⁻⁴, α_s = −9.7×10⁻⁴, N* = 50.37 at C = 8×10⁻⁷. Untilted, n_s = 1 − 2/N − 3/((D−2)N²) and r = 8/((D−2)N²), where D is the dimension of the effective theory at horizon exit; a locked run with D = 6 gives r = 1.07×10⁻³ (exact MS) at the same n_s, so a future r measurement reads off D. Across thirteen dimensional reductions the string-coupling direction is the only one carrying the plateau. Open steps and relations to open questions are stated in the paper. Status of the claims. Species-gradient principle: postulate. Identity, plateau coefficient, end point and r(D): derived and checked numerically. Tilt: estimate (BBHL form). Predictions: exact Mukhanov–Sasaki, all runs pre-registered (locks in the repository). This model was formulated after the CMB-SPA and DESI data and is scored only on future data; the SERAPIS-E predictions frozen on 15 Nov 2026 are unchanged. Not peer reviewed. License: CC BY-NC-SA 4.0 under the AEGIS Community License — free for individuals, scientists and students; any business or for-profit use needs the AEGIS Commercial License (https://github.com/dannyeldridge54/AEGIS/blob/master/licensing/COMMERCIAL_LICENSE.md). Built by one independent researcher with the help of AI. Code and pre-registration files: https://github.com/dannyeldridge54/AEGIS (branch unify-bg; notes D1527–D1606). Prepared with an AI assistant (Claude, Anthropic).
Language: English
Open Access status: green
OpenAlex ID: W7236825017
Cites:
Cited by: 3
Related to: 9