Answers are grounded in this paper and cite their sources.
Documentation-only v2.0 methodological note in the existing Σ-AFO/HM version family. This original English PDF derives a bounded adaptive-routing rule in which human authority, evidence and risk are mandatory gates rather than compensable scores. Nineteen local finite synthetic tests passed. They do not establish general agent safety, measured time, token, energy or monetary savings, independent assessment of authority controls, or safe external deployment. The historical logical-cost report is kept separate and is not re-labelled as a physical resource measurement. Only this explanatory PDF is uploaded; no earlier code, archive or third-party figure is included. Earlier version: https://doi.org/10.5281/zenodo.22136592 . CC BY 4.0 applies only to this newly original PDF, not to earlier software or cited works. Computational and AI assistance is disclosed; Riccardo Giudici is responsible for the presentation as an independent systems orchestrator. No institutional endorsement is claimed.
No takes yet. Share an insight, caveat, or question.
Riccardo Giudici (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: