Glossary of AI Nativeness

AI Nativeness is the degree to which an organisation or individual has integrated AI in ways that build rather than deplete human capability. AI Nativeness is distinct from AI adoption or AI readiness: it is not about whether AI tools are in use, but whether that use is creating compounding advantage or compounding disadvantage.

Full explanation, how it differs from AI-native and AI readiness, and how it is measured: What is AI Nativeness?

See also: Foundation Paper · Compounding Advantage · AI Dependency

AI dependency is a pattern of AI use in which human capability atrophies because AI is performing tasks that would otherwise develop and sustain human skill. AI dependency is the opposite of AI Nativeness. It typically manifests gradually: short-term productivity gains mask a steady erosion of the human capital on which durable performance depends.

Dependency is architecturally different from adoption. A dependent individual or organisation produces reasonable outputs in the short term while becoming structurally less capable over time.

See also: AI Deskilling · Substitution · Compounding Disadvantage

AI deskilling is the process by which repeated AI substitution erodes a specific skill or domain of human competence. Deskilling occurs when AI performs tasks that would otherwise develop and sustain human capability. It can occur at the individual level (a professional who no longer writes, analyses, or decides independently) or at the organisational level, where institutional knowledge is lost because AI summarises everything and humans stop building understanding.

Full explanation, recent research and how to measure it: AI deskilling.

See also: AI Dependency · Augmentation · Kaivant-I

AI readiness is the extent to which an organisation has the infrastructure, governance, and processes to deploy AI. AI readiness measures are widely used by analysts, consultancies, and regulators. They are not designed to show whether AI use is building or eroding the human capability that performance depends on.

An organisation can score at the top of every AI readiness framework while actively deskilling its workforce. Readiness tells you whether you can deploy AI; it tells you nothing about whether that deployment is building or eroding the human capability on which performance ultimately depends.

See also: AI Nativeness · Foundation Paper

Compounding advantage is the dynamic in which strong AI integration produces increasing returns: human capability and AI capability reinforce each other, producing outputs and adaptability that neither could achieve independently. Organisations and individuals with high AI Nativeness scores are structurally positioned for compounding advantage.

The compounding effect operates over time: early investment in augmentation-oriented AI use widens the capability gap between AI-native and AI-dependent actors.

See also: Compounding Disadvantage · The Kaivant Score

Compounding disadvantage is the mirror dynamic to compounding advantage, in which AI dependency erodes the human capability required to deploy AI effectively. High AI dependency scores signal compounding disadvantage risk: a spiral in which increasing reliance on AI produces decreasing genuine performance, leaving the organisation or individual progressively less capable of operating without it.

See also: Compounding Advantage · AI Dependency

Augmentation is an AI use pattern in which the person develops or retains capability through the interaction. In augmentation, AI functions as a leverage multiplier: the person makes better decisions, produces more sophisticated outputs, or learns faster through AI assistance, while the underlying human skill remains intact or grows.

Augmentation requires active human engagement with the AI process, not passive consumption of AI output. The distinction between augmentation and substitution is often not visible in the output. It is visible in what the person can do independently after the interaction.

See also: Substitution · Kaivant-I · AI Deskilling

Substitution is an AI use pattern in which the person defers cognitive or creative work to AI in ways that prevent skill formation or cause skill atrophy. Substitution produces short-term output quality but long-term capability decline. It is the primary driver of AI deskilling at the individual level.

Substitution is not always a conscious choice. Many AI interfaces are designed to make substitution the path of least resistance: the output appears immediately, the cognitive work is optional. Kaivant-I measures the patterns that accumulate from these choices over time.

See also: Augmentation · AI Deskilling · Human Capital Depth

Personal Leverage Architecture is one of two axes in the Kaivant-I assessment, measuring how effectively an individual uses AI for productive leverage, across Personal Workflow Automation Rate, Personal Decision Velocity, Coordination Footprint and Personal Leverage Trajectory. Personal Leverage Architecture is primarily a lagging indicator: it reflects what AI use has produced so far.

High Personal Leverage Architecture alongside declining Human Capital Depth is the primary signal of substitution risk: the individual is producing well today but eroding the capability on which future performance depends.

See also: Human Capital Depth · Kaivant-I

Human Capital Depth is one of two axes in the Kaivant-I assessment, measuring the depth and resilience of an individual's underlying human capabilities: Judgement Independence, Cognitive Range, Relational Intelligence, and Capability Independence and Resilience. Human Capital Depth is primarily a leading indicator of long-term performance.

A declining Human Capital Depth score alongside high AI use is the central deskilling signal in the Kaivant-I framework. An individual can maintain output quality while Human Capital Depth erodes; the divergence is what makes the assessment structurally different from productivity metrics.

See also: Personal Leverage Architecture · Substitution · AI Deskilling

The non-compensatory composite is the scoring logic used by both the Kaivant-O and Kaivant-I assessments. It does not allow a strong score on one axis to offset a weak score on the other. The Kaivant Score floors at the weaker of the two axes: an organisation that is excellent at deploying AI leverage but actively eroding its Organisational Capital receives a score reflecting the weaker axis, not the average.

This design reflects a structural insight: capability gaps are not cancelled by capability strengths elsewhere. A highly automated organisation with collapsing human judgement is not a balanced organisation. It is a fragile one.

See also: The Kaivant Score · Foundation Paper

The two-layer framework architecture is how the KAIVANT framework is built: two layers with different durability and governance. The conceptual layer (three claims: the two-axis compounding thesis with its non-compensatory composite, the augmentation-substitution distinction, and the managed loop) is designed for 10-20 year durability and requires formal revision with published rationale. The indicator layer (specific metrics, thresholds, and benchmarks) is reviewed annually and versioned with a published change log.

The distinction ensures the framework can evolve with the evidence base without losing definitional stability. What AI Nativeness is does not change; how it is best measured is expected to improve.

See also: Foundation Paper · Validation programme