2 cumulative citations
View corpus contextEuropean regions with stronger innovation systems and greater AI adoption tend to grow faster — R&D intensity, internet coverage and ICT expansion align with better regional economic performance; regions lagging in AI and smart specialisation risk falling further behind.
An Attempt to Assess the Impact of AI as a Modern Tool for Regional Policy in the Process of Innovative and Economic Development of European Regions
openalex
correlational
low evidence
7/10 relevance
Summary only
summary available; pdf_status=paywall
DOI
Source
PDF
Across European NUTS2 regions, stronger regional innovation systems—marked by higher R&D spending, greater internet coverage, ICT-sector growth and indicators of AI implementation—are associated with higher regional economic development.
Citation observations
Cumulative provider counts captured on specific dates; providers are never combined.
2 cumulative citations
View corpus contextThe beginning of the 21st century is associated with a significant technological leap on a global scale, which has had a substantial impact on production and economic processes at the national and regional levels. This radical technological change in the economy is linked to the emergence and development of artificial intelligence and effective knowledge management, which are the main drivers of economic growth. The use of AI today can be traced in many different areas of applied science—medicine, physics, mathematics, and engineering design, including modeling, planning, and management of territorial systems. The accumulation of large databases and other information necessary for AI to function is directly related to the spatial aspects of economic development, which is also based on local potential (a place-based approach). At the same time, local knowledge resources and innovation potential are not fully utilized in the context of technology diffusion and AI implementation in individual countries and regions. In this regard, this study aims to analyze the role of regional innovation systems, with a focus on AI development, and to track their impact across individual European regions, using NUTS2 spatial-level data to ensure objectivity. The authors consider AI innovation a modern tool for decision-making in the implementation of regional policy, with a specific impact on cohesion between EU regions. The results of the study show a direct link between the localization of regional innovation systems, R & D expenditure, AI implementation, and the economic development of European regions. Important factors influencing this process are the degree of Internet coverage, the capacity to generate innovation, the degree of AI implementation in the individual economic sectors of the countries, the growth of the ICT sector in relation to the overall development of GDP and the economy, and the result of the smart specialisation of regional innovation systems.
Summary
Main Finding
There is a direct, positive relationship between the localization and strength of regional innovation systems (RIS) — proxied by R&D expenditure, AI implementation, and smart specialisation — and economic development across European NUTS2 regions. Factors such as Internet coverage, innovation-generation capacity, sectoral AI adoption, and ICT-sector growth amplify this effect and shape regional convergence/cohesion within the EU.
Key Points
- AI and knowledge management are central drivers of 21st-century technological change, affecting production and regional economic processes.
- Spatial aspects matter: accumulation of the data and information that power AI is tied to place-based local potential and RIS.
- Local knowledge resources and innovation potential are often underutilized in national and regional technology diffusion and AI uptake.
- Using NUTS2-level analysis, the study links RIS localization, R&D spending, and AI implementation to improved regional economic outcomes.
- Critical mediating factors include:
- Degree of Internet coverage (digital infrastructure and access).
- Capacity to generate innovation (local R&D and innovation ecosystems).
- Extent of AI implementation across economic sectors (sectoral diffusion).
- Growth of the ICT sector relative to GDP (digital industry dynamics).
- Results of smart specialisation strategies within RIS (targeted regional strengths).
- The authors argue AI can be a decision-making tool for regional policy and has specific implications for EU regional cohesion.
Data & Methods
- Spatial scope: European regions at the NUTS2 level (used to ensure objective, comparable regional analysis).
- Core variables/indicators discussed: R&D expenditure, measures of AI implementation, Internet coverage, ICT sector growth, indicators of innovation capacity, and smart specialisation outcomes.
- Analytical approach (as described): comparative, region-level analysis tracking links between RIS characteristics and regional economic performance. (The paper emphasizes spatial, place-based measurement using NUTS2 units; specific econometric techniques are not detailed in the provided summary.)
- Implicit methodological elements: assembly and cross-regional comparison of innovation, digitalization and economic indicators; likely use of regional statistics and indicators to assess correlations/associations between AI-related variables and economic outcomes.
Implications for AI Economics
- Place-based policy is essential: regional heterogeneity in data-assets, skills, and RIS means AI-driven growth requires tailored regional strategies rather than one-size-fits-all national policies.
- Infrastructure and access matter: expanding Internet coverage and digital infrastructure is foundational for AI diffusion and equitable regional growth.
- Invest in local innovation capacity: boosting regional R&D and strengthening RIS will increase the ability to generate and apply AI-driven innovations.
- Sectoral adoption focus: policies should encourage AI uptake across diverse economic sectors, not only in ICT hubs, to broaden benefits and foster cohesion.
- Align smart specialisation with AI: regionally targeted specialisation strategies that incorporate AI can amplify comparative advantages and productivity gains.
- Cohesion and redistribution: EU/regional policymakers should account for the risk of divergence from uneven AI adoption and direct cohesion funds or programs to build capacity in lagging regions.
- Data and governance: because AI depends on large local datasets, policymakers should support data ecosystems, data-sharing frameworks, and governance that enable safe, effective regional AI use.
- Research priorities: further work should quantify causal impacts (e.g., via panel or spatial econometrics), identify spillover channels between regions, and evaluate which policy levers most cost-effectively accelerate inclusive AI diffusion.
Assessment
Paper Typecorrelational
Evidence Strengthlow — The study appears to rely on cross-sectional or aggregated regional correlations using NUTS2 data without a clear quasi-experimental design, instrument, or other strategy to address endogeneity, reverse causality, and omitted variable bias, so causal claims are weak.
Methods Rigormedium — Using NUTS2 regional data and standard indicators (R&D expenditure, internet coverage, ICT share, etc.) is appropriate for describing spatial patterns and associations and may permit multivariate and spatial controls, but the methods as described do not demonstrate strong identification, robustness checks, or detailed measurement validation for 'AI implementation' proxies.
SampleAggregated European regions at NUTS2 spatial level (EU regions), using regional indicators such as R&D expenditure, proxies for AI implementation across sectors, internet coverage, ICT-sector growth, GDP/economic output and smart specialisation metrics; time period not specified in the abstract and sample size not reported.
Themesinnovation adoption
GeneralizabilityFindings are specific to European NUTS2 regions and may not generalize to non-European countries or subnational units with different institutional contexts., Aggregate regional analysis risks ecological fallacy and cannot identify firm- or worker-level mechanisms., Measurement of 'AI implementation' likely relies on imperfect proxies (sectoral composition, ICT activity), limiting inference about true AI diffusion., Cross-sectional/aggregate approach may conflate long-run structural differences with contemporaneous associations, limiting temporal generalizability.
Claims (9)
| Claim | Direction | Outcome | Confidence & Evidence | Details |
|---|---|---|---|---|
| The beginning of the 21st century is associated with a significant technological leap on a global scale, which has had a substantial impact on production and economic processes at the national and regional levels. Fiscal And Macroeconomic | positive | production and economic processes at national and regional levels |
Reading fidelity
high
Study strength
low
|
not reported
|
| This radical technological change in the economy is linked to the emergence and development of artificial intelligence and effective knowledge management, which are the main drivers of economic growth. Fiscal And Macroeconomic | positive | economic growth (linked to AI and knowledge management) |
Reading fidelity
high
Study strength
speculative
|
not reported
|
| The use of AI today can be traced in many different areas of applied science—medicine, physics, mathematics, and engineering design, including modeling, planning, and management of territorial systems. Research Productivity | positive | presence/use of AI across applied science fields |
Reading fidelity
high
Study strength
low
|
not reported
|
| The accumulation of large databases and other information necessary for AI to function is directly related to the spatial aspects of economic development, which is also based on local potential (a place-based approach). Adoption Rate | positive | relationship between data accumulation (large databases) and spatial economic development / local potential |
Reading fidelity
high
Study strength
low
|
not reported
|
| Local knowledge resources and innovation potential are not fully utilized in the context of technology diffusion and AI implementation in individual countries and regions. Innovation Output | negative | utilization of local knowledge resources and regional innovation potential |
Reading fidelity
high
Study strength
medium
|
not reported
|
| This study analyzes the role of regional innovation systems, with a focus on AI development, and tracks their impact across individual European regions using NUTS2 spatial-level data to ensure objectivity. Adoption Rate | null_result | impact of regional innovation systems (with focus on AI) across European regions |
Reading fidelity
high
Study strength
medium
|
not reported
|
| The authors consider AI innovation a modern tool for decision-making in the implementation of regional policy, with a specific impact on cohesion between EU regions. Governance And Regulation | positive | impact of AI-based decision-making on regional policy and EU regional cohesion |
Reading fidelity
high
Study strength
speculative
|
not reported
|
| The results of the study show a direct link between the localization of regional innovation systems, R & D expenditure, AI implementation, and the economic development of European regions. Fiscal And Macroeconomic | positive | economic development of European regions |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Important factors influencing this process are the degree of Internet coverage, the capacity to generate innovation, the degree of AI implementation in the individual economic sectors of the countries, the growth of the ICT sector in relation to the overall development of GDP and the economy, and the result of the smart specialisation of regional innovation systems. Fiscal And Macroeconomic | positive | influence on the link between regional innovation/AI and regional economic development |
Reading fidelity
high
Study strength
medium
|
not reported
|