The DIG Framework

Overview

The DIG framework turns a one-line definition into something an organization can act on. It names the four things that have to be true for an AI-influenced decision to be defensible, and gives each one a place in the governance program.

Pillar 1: Information Provenance

Every decision rests on information. Provenance asks where that information came from and whether it can be trusted: the source, the chain of custody, and the freshness. Without provenance, a decision cannot be reconstructed or defended.^[1]

Pillar 2: Decision Traceability

Traceability is the record of the decision itself: what was recommended, by which system, on what basis, who reviewed it, and on what authority it was acted on. It is the difference between "the AI did it" and "a named person decided, and here is the trail."

Pillar 3: Representation Integrity

AI systems now describe your company to the outside world. Representation integrity is keeping that description accurate across AI search, models, and data environments, because a misrepresentation becomes evidence a regulator, partner, or court can cite. See Representation Integrity.

Pillar 4: Audit Readiness

Audit readiness is the ability to prove, on demand, that the first three pillars held. It is the posture that turns "we have AI policies" into "we can show our oversight worked."

How the pillars map to regulation

The four pillars are not arbitrary. They correspond to controls described across the major AI governance regimes, where most obligations overlap.