Work

The Governance–adoption asymmetry of artificial intelligence: A systematic review of educational, labour, and cybersecurity policy gaps

Year: 2026
Type: review
Study design: Systematic Review
Abstract: This systematic narrative review compares the pace of artificial intelligence (AI) adoption with institutional readiness to govern it in education, employment and cybersecurity. It has two objectives: to synthesise the 2017–2026 evidence base and to develop an exploratory framework for cross-sector comparison. The Governance–Adoption Asymmetry (GAA) Framework defines a core asymmetry index, Δ, as the gap between Adoption Velocity and Governance Readiness, and treats three further dimensions (Distributional Inequity, Evidence Uncertainty and the Augmentation–Displacement Balance) as conditions that shape the consequences of that gap. All dimensions are scored on a transparent 1–5 anchor rubric in which higher values denote higher risk; the five-dimension mean is reported separately as a broader GAA Risk Profile. Drawing on 68 sources retained through PRISMA 2020-aligned screening of 412 records, the core asymmetry is widest in cybersecurity (Δ = +3.0 on a −4 to +4 scale; Risk Profile 3.8/5), followed by education (+2.0; 3.2/5) and employment (−0.5; 2.7/5). The Risk Profile ranking holds under 95.8% of 100,000 random weightings, although education and cybersecurity tie under median aggregation. All scores are single-coder interpretive judgements, not validated measurements. AI tutoring yields gains in controlled trials while institutional AI policy coverage remains low; estimates of technical automation potential diverge from observed employment stability; and AI-enabled attacks are spreading faster than AI-specific security governance. Seven actor-specific policy scenarios are offered as normative propositions for testing.
Language: English
Open Access status: gold
Estimated APC paid: $1,180
OpenAlex ID: W7160283071
FWCI:
Cites: 18
Cited by:
Related to: 10