The Architecture of Proof Glossary: Precise Definitions | AI Governance

A standard metric for model accuracy and separability. In the Architecture of Proof, AUC is considered a "Stage 1" metric—it provides a population-level aggregate that can mask individual-level logic failures.

A visualization that maps predicted probabilities to actual frequencies. A well-calibrated model ensures that when it says there is an 80% chance of fraud, fraud occurs exactly 80% of the time in the long run.

The "silent killer" of autonomous systems. Causal drift occurs when the logical relationship between an input and an output shifts, even if the data distributions (covariates) remain stable.

A "black box flight recorder" for AI logic. Unlike a standard log file that shows what happened, a Causal Trace captures the step-by-step logical path from data ingestion to final output, including which specific business rules or model layers carried the most weight.

A four-level hierarchy that dictates the "leash length" of an AI system, ranging from Tier 1 (Observe) to Tier 4 (Full Autonomy).

The general term for the decay of model performance over time as production data diverges from training distributions.

In the Architecture of Proof framework, Evidence-Based Promotion is the formal governance bridge between a "lab-safe" pilot and "high-stakes" production autonomy. Rather than promoting an AI system based on static accuracy scores or arbitrary project deadlines, this protocol requires a system to "earn" its way into higher Control Tiers by demonstrating a dense, replayable audit trail.

A system is ready for promotion only when its Proof Infrastructure—including its Benford Perimeter for data integrity and its Causal Trace for logic transparency—is robust enough to handle the inevitable 8% of cases where the probabilistic model will fail. By anchoring promotion in defensibility rather than just prediction, organizations transform AI from a black-box liability into a high-ROI, auditable business asset.

Local Interpretable Model-agnostic Explanations. A post-hoc technique that learns a "surrogate" linear model around a single prediction to provide local intuition.

The technical capability to reconstruct the exact state of an AI system at the millisecond a specific decision was made. This includes the model version, precise input data (feature vector), external API responses, and the specific rules or weights that fired.

Shapley Additive Explanations. A game-theoretic approach to explaining model outputs by assigning an "importance" value to each feature based on its contribution to the final prediction.

The convergence of probabilistic model patterns into deterministic-like consistency. Stable Reasoning occurs when an AI system produces non-divergent logical paths (Causal Traces) for similar inputs, eliminating the "stochastic variance" that typically characterizes unmanaged LLMs.