The DBS Standard v0.1: An Investigative Methodology for Compromised Case Files and AI-Assisted Epistemic Systems

The DBS Standard (v0.1) is a replicable investigative journalism methodology designed specifically for complex or historical cold cases where the traditional investigative file is compromised, managed, or incomplete. Traditional investigative journalism frequently treats police reports, institutional files, and official records as a neutral starting point. However, in cases involving institutional corruption, systematic non-resolution, or historical evidence destruction, relying primarily on the official file traps subsequent investigations within a tightly managed narrative. Developed during the multi-decade cold case murder of civil rights figure and attorney Alberta Odell Jones (1930–1965) in Louisville, KY, this framework establishes a system to "build below the file". By constructing a baseline narrative entirely from independent, public, and open-source datasets before analyzing official case documents, the researcher can map institutional gaps and surface hidden connections. Core Framework Components The methodology introduces eight interlocking operational components to ensure systemic rigor and epistemic safety: The Epistemic Ledger: A three-tier classification system (Documented, Inferred, Speculative) that enforces strict verification boundaries and prevents theory from outrunning evidence. Node Graph Architecture: A non-linear network database design that catalogs individuals, locations, timelines, and themes to distinguish between "radiating" and "bridging" entities. Sprint Structure: Exhaustive, question-driven research cycles that map breakthroughs, dead ends, and course corrections iteratively. Property-First Lens: A deliberate inversion of investigative sequence focusing on deed chains, probate records, and material beneficiaries over state-provided motives. Institutional Rotation Mapping: Tracking personnel movements, timeline correlations, and professional career anomalies within key oversight roles. Architectural & Infrastructure Mapping: Evaluating the design, funding, and governance of the physical environments and record-keeping repositories linked to a case to identify curated archive gaps. Counter-File Independence: Strict operational security rules regarding framework insulation, prior investigator engagement timelines, and signal containment. Records Pull List: A prioritized, risk-rated, location-organized field log that translates open speculative questions into actionable courthouse and archive research. The Watson Model (AI-Assisted Integration) The DBS Standard outlines an explicit operational role for Large Language Models (LLMs) like Anthropic's Claude Code as formal research partners. Operating inside a human-controlled epistemic system, the AI is restricted from generating conclusions or classifying evidence. Instead, it functions as a localized network pressure-tester to uncover structural discrepancies, map data anomalies, and generate verifiable hypotheses that must pass independent validation before publication.