The G(I,D,t) Coherence Framework
Overview
A forensic framework for data integrity and ethical AI governance. The Digital Integrity Foundation is developing G(I,D,t), a candidate coherence metric introduced here as a diagnostic instrument that requires independent replication, not as a settled scientific finding. The framework pairs concrete technical safeguards with institutional governance so that AI systems can be trusted because they can be examined, not merely because they are asserted to be trustworthy.
Core Research Commitments
Our work rests on a few simple commitments: truth before convenience, evidence over opinion, falsifiability over dogma, transparency over opacity, and service to the people and institutions our work is meant to protect. We treat measurement as a discipline of humility. We do not claim perfection; we measure how closely systems honor the structures that sustain coherence, and we report what we find openly and with the limits clearly stated.
The Pythagorean Harmonic Audit
The framework explores whether harmonic ratio analysis can serve as a diagnostic lens for structural coherence across complex systems. The approach is organized in two layers. The first layer is a laboratory of independent, unit-consistent metrics drawn from verifiable instruments and records. The second layer is a harmonic audit that applies ratio analysis to the core residuals, with the aim of detecting coherence or dissonance under stress. Technical work is kept strictly separate from philosophical and cultural exploration, so that evidentiary outputs are never mixed with interpretive narrative.
Data and Attribution Notice
The Digital Integrity Foundation develops measurement instruments for coherence in complex systems. The G(I,D,t) framework is a candidate metric under active development, and the Architecture of Integrity is a governance scaffold designed for practical deployment.
The Foundation applies its own metrics to publicly available datasets. Datasets produced by NASA’s PUNCH mission, the sPHENIX collaboration, and other public research efforts remain the work of their originating institutions. Any application of DIF’s metrics to those datasets is the Foundation’s own independent analysis and does not constitute or imply validation, endorsement, or confirmation by those institutions.
All empirical claims are subject to independent replication. A pre-registered falsification protocol, specifying the conditions under which the framework would be revised or abandoned, is in preparation for deposit with the Open Science Framework.
What This Work Is, And What It Is Not
It is a forensic audit of coherence across independent systems, a diagnostic tool for stress and integrity assessment, and a method designed to be transparent, falsifiable, and reproducible. It is not a claim about consciousness, awareness, or any non-physical phenomenon. It is not a replacement for domain expertise or governance, and it is not a scoring system for people or identities. Metaphors such as a stethoscope for integrity illustrate purpose; they are not scientific claims.
Our Values
Truth over convenience. Measurement with humility. Transparency by design. Stewardship for future generations. Unity in diversity, harmony in action. These principles guide both our technical work and our public commitments.
Publications in Preparation
The framework is documented in the Foundation’s white paper, Architecture of Integrity: A Forensic Framework for Ethical AI Governance. Technical specifications, including the reference cadence, bandwidth parameters, and reporting bands, are documented in internal technical memos and remain subject to ongoing refinement and independent replication.
Digital Integrity Foundation | Canadian Federal Non-Profit Corporation No. 1591720-4 | 607-16 McAdam Avenue, Toronto, ON M6A 0B9, Canada