Competitive Intelligence Capability and Strategic Innovation Performance: Evidence from Global AI Patent and Knowledge Transfer Networks
Abstract
Purpose:This study examines how competitive intelligence (CI) capability supports strategic innovation performance in frontier artificial intelligence industries by transforming patent, publication, and semantic knowledge-transfer signals into actionable intelligence for strategic decision-making. Methodology/Approach: A quantitative secondary-data design was applied to the DeepInnovationAI dataset, comprising 2.3 million patents and 3.5 million scientific publications covering the period 1960–2020, with the main analysis focused on 987 country-year observations from 47 countries during 2000–2020. Multiple linear regression, mediation analysis, Pearson correlation, temporal trend decomposition, and k-means clustering were used to evaluate how Patent Novelty Index, Knowledge Transfer Rate, R&D Intensity, temporal velocity, and publication volume contribute to Competitive Intelligence Score. The empirical model is repositioned as an open-source technology intelligence framework that maps innovation signals onto the CI cycle of environmental scanning, signal analysis, dissemination, decision use, and feedback. Findings: Patent Novelty Index is the strongest predictor of Competitive Intelligence Score (beta = 0.41, p < 0.001), followed by Knowledge Transfer Rate (beta = 0.33, p < 0.001). The regression model explains 64% of the variance in Competitive Intelligence Score. Knowledge Transfer Rate partially mediates the relationship between R&D Intensity and CI Score, with an indirect effect of beta = 0.11. Cluster analysis identifies three national CI profiles: Established Leaders, Emerging Challengers, and Peripheral Contributors. Originality/Relevance: The study contributes to the Journal of Sustainable Competitive Intelligence by moving beyond descriptive patent and bibliometric analysis and operationalizing CI as a strategic capability. It incorporates CI architecture, intelligence governance, intelligence dissemination, strategic foresight, and decision-support logic into the interpretation of AI innovation data. This framing is aligned with recent journal contributions that conceptualize AI-enabled CI as a strategic capability and emphasize CI architecture as a foundation for strategic decision-making.Theoretical/methodological contributions: The study develops a measurable CI-cycle framework for frontier technology analysis by linking patent novelty, knowledge transfer, temporal velocity, and geographic clustering to intelligence planning, environmental scanning, analysis, dissemination, decision support, and feedback. Methodologically, it demonstrates how large-scale open innovation datasets can be used as strategic intelligence infrastructures, while clearly defining the 2000–2020 period as a pre-generative-AI empirical baseline.