Glossary

Intelligence layer

The intelligence layer is the engineered surface where an organization turns live signals into coherent action: sensing, interpretation, decision logic, execution, and feedback — repeatedly, under imperfect information.

Why it matters

Bolt-on AI features (copilots, recommendations, automation widgets) are easy to copy and often plateau because they sit at the product surface without a durable decision system underneath. What compounds is the layer that decides what to notice, what to ignore, what to prioritize, and what to do next — then revises itself through correction.

“Intelligence layer” is TSS terminology for this decision-system concept. Related essays argue why features fail to compound; Knowledge owns the concise definition so topics and Discover bridges stay consistent.

Mechanisms

A decision stack binds sensing → interpretation → decision logic → execution → feedback into one mechanism rather than separate departmental artifacts.
Products act as decision surfaces: interfaces generate correction naturally through hesitation, overrides, abandonment, and edits — not only through dashboards.
Governable decision logic encodes tradeoffs (speed vs accuracy, growth vs trust) so outputs can be audited, constrained, escalated, or killed at runtime.

Boundaries

An intelligence layer is not synonymous with “having a model” or “having analytics.” Prediction without decision rights and write-back is observation, not judgment.
Signal rot, latency creep, incentive drift, and governance lag can make the layer look active while judgment quietly degrades.
Do not equate the intelligence layer with a marketing “AI feature roadmap”; Knowledge uses the term for system design, not feature checklists.