Pre-registered standing prediction: the value of the fine-structure constant (sealed, Tier 1)

Read full text
Authors: Gabriel Steiger
Cited by: 2
References: 3

Abstract

WHAT BECOMES PUBLIC IN THIS DEPOSIT, EXACTLY: a 13-digit numerical value of the inverse fine-structure constant, falsification/corroboration clauses, and SHA-256 hash commitments. NO formula, NO mechanism, NO geometry, and no statement of the construction beyond 'parameter-free' are disclosed; the construction is hash-committed and held privately. PRE-REGISTERED STANDING PREDICTION: THE VALUE OF THE FINE-STRUCTURE CONSTANT (sealed, Tier 1 -- deposited 2026-06-12) Author: Gabriel Steiger (independent researcher) Protocol: this deposit follows the author's seal-then-measure protocol (cf. 10.5281/zenodo.20360004, 10.5281/zenodo.20360259), including one completed cycle in which sealed predictions failed and were reported with full disclosure per the seal's own terms (seal: 10.5281/zenodo.20626679; falsification report: 10.5281/zenodo.20651313). The construction behind the present prediction is fully documented in private files whose SHA-256 hashes appear below; it contains NO continuous adjustable parameters. The prediction is exact and permanent: falsifiable by every future improvement in the measurement of alpha, indefinitely. 1. THE PREDICTION The inverse fine-structure constant has the exact value alpha^-1 = 137.0359991718033 ... (13 significant digits shown; the full-precision value -- an exact algebraic specification -- is contained in the hashed document HV below.) Reference: CODATA 2022 gives alpha^-1 = 137.035999177(21); the sealed value sits at -0.038 ppb (-0.25 sigma). The content of this seal is not the present agreement but the STANDING CLAIM: as the experimental uncertainty shrinks -- by any future method, to any future precision -- the recommended value converges to the digits above and to their hashed continuation. 2. FALSIFICATION AND CORROBORATION - FALSIFIED if any future CODATA recommended value (or any direct determination subsequently confirmed by an independent method) differs from the sealed value by more than 5x its stated standard uncertainty. - CORROBORATED each time a recommended value with smaller uncertainty than the previous one remains within 3x its uncertainty of the sealed value. - The author commits to reporting either outcome, as in 10.5281/zenodo.20651313. 3. SHA-256 COMMITMENTS (private documentation, frozen 2026-06-12) [HV] VALUE -- full-precision value + exact algebraic specification: ad3a7b05c97a2b28277045d0259403d37bf238b9417769c84530fba50905687b [HA] DOC-A -- the construction paper (complete): 9ad6064811a6fdea53ee31a3667b2f05a23d70b801c95f0ac748348f4dce33e7 [HB] DOC-M -- the mechanism record (eleven documents, zip): 50b3719e9ca7da9acce3f4337c9cc5b26950e7ab2eeb34f69dfd543b381ffe09 [HC] DOC-S -- scripts and machine outputs (zip): 824cfa55b3495de7ae7af03bec5703173cc1f41857668a50fa6e3c95be257393 4. DISCLOSURE COMMITMENT The construction will be disclosed in full upon either (i) falsification per section 2, or (ii) the author's election, whichever comes first. Questions of interpretation resolve against the hashed documents.

Fields of Study & Subjects

Scientific Measurement and Uncertainty Evaluation Machine Learning in Materials Science Scientific Computing and Data Management

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
12 June 2026