Relocating Trust to the Verifier: A Truth-Maintenance System for an AI-Generated Theory of Everything

Abstract

The runnable companion to the methodology that produced the canon above. A fluent, approval-seeking LLM pointed at an open-ended theory-of-everything by a researcher who wants it to succeed forms a mutual-confirmation loop with no reality-check on either side — the maximally dangerous case, because the mathematics can always be made internally consistent and the one decisive check (experiment) is decades away or absent. The cure is structural rather than a better model: relocate trust from the generator to a verifier. PTMS is a truth-maintenance system that enforces consistency, not truth — it inverts the classical TMS assumption that justifications are ground truth, treating every AI-supplied justification as untrusted until it binds to checkable evidence: a cited script that exits 0, a retired claim's signature absent at every surviving site, a cross-reference that resolves. Three layers of increasing cost — check (read-only consistency over the canon), verify (re-executes the cited self-asserting scripts), and graph (resolves the cross-reference dependency graph) — run over a machine-readable sidecar registry that never edits the human-facing ANCHOR.md / DRIFT.md prose. Reports its behaviour on a real six-month AI-assisted physics corpus, mechanically catching named failure classes: un-propagated multi-part retractions, evidence regressions, seductive numerical coincidences, and live claims standing on retracted foundations. Explicit about the limit — the tool makes such research auditable and constrained, not correct; only forward prediction closes the remaining gap, and no apparatus can manufacture it.

Keywords

How to cite

Elliman, D. (2026). Relocating Trust to the Verifier: A Truth-Maintenance System for an AI-Generated Theory of Everything. Neuro-Symbolic Ltd technical report. https://doi.org/10.5281/zenodo.20786144

@techreport{elliman2026ptmspaper,
  author      = {Elliman, David},
  title       = {Relocating Trust to the Verifier: A Truth-Maintenance System for an AI-Generated Theory of Everything},
  institution = {Neuro-Symbolic Ltd},
  year        = {2026},
  doi         = {10.5281/zenodo.20786144},
  url         = {https://neusym.ai/papers/ptms_paper}
}