Semantic Drift Detector
Ki · Single Ability
The Problem
The term 'machine learning' now includes all AI methods, showing assumes definition stability over time.
The Operation
The model is directed to anchor key terms with operational definitions: what each includes, excludes, how to test. At each reasoning stage, re-extract working definitions and compare to anchors. Flag semantic drift: scope creep, narrowing, or meaning substitution. On silent drift, force re-anchoring: restate original, state divergence, choose one definition. Verify no conclusion rests on equivocation. Reason ambiguous terms is rejected.
The Structure
This ability runs on an anchor-drift-correct cycle that detects when reasoning has drifted from its reference point. The monitor runs continuously, checking for drift at each step.
If a key term is used later in the analysis with a subtly different meaning than when it was introduced, definitional consistency tracking was not performed.
Haki · Multi Ability
Synergy Topology
In Haki mode, the API retrieves the primary ability first, then fans out to three synergy roles that compound its reasoning.
When retrieved in Haki mode, these four abilities don't run in sequence. They merge into a single injection where the dependency grounds the reasoning context, the amplifier sharpens the primary's output, and the alternative provides a fallback path if the primary's topology cannot converge. The result is a multi-angle reasoning scaffold that covers failure modes no single ability can reach alone.
Appears in Use Cases