Building institutional resilience to AI-driven misinformation in critical infrastructure: Evidence from the Albanian energy sector

Abstract

The rapid expansion of generative artificial intelligence has significantly transformed digital information environments, increasing the volume, velocity, and technical sophistication of misinformation affecting critical infrastructure systems. In the energy sector, such disruptions pose measurable risks to regulatory reliability, infrastructure investment, and operational stability, particularly in countries undergoing energy transition and digital modernization. Albania’s developing energy ecosystem provides a relevant empirical context for examining these emerging system-level vulnerabilities. This study analyzes the integration of Business Intelligence (BI), Competitive Intelligence (CI), and AI-based detection systems in strengthening institutional resilience against AI-generated misinformation. A qualitative-dominant mixed-methods case study approach is employed, combining large-scale digital media monitoring, intelligence-cycle modeling, and expert-based validation. The research focuses on a coordinated disinformation campaign targeting the Qeparo Solar Farm project in 2025, using temporal network analysis, content classification, and attribution mapping to evaluate diffusion dynamics and institutional response mechanisms. Results indicate that misinformation propagation followed structured temporal patterns, emotionally optimized framing strategies, and coordinated amplification networks consistent with organized influence operations. Early anomaly detection was achieved through hybrid analytical systems integrating automated machine learning tools with professional assessment. Competitive Intelligence analysis supported probabilistic attribution and risk prioritization, while coordinated governance responses enabled rapid system stabilization and restoration of stakeholder confidence. The study proposes an applied governance-oriented resilience framework integrating BI, CI, and AI detection within a unified institutional monitoring architecture. The findings demonstrate that effective protection of critical infrastructure information systems depends primarily on institutional system design, operational coordination, and analytical capacity, rather than technological deployment alone. This research provides practical guidance for regulators, engineers, and infrastructure managers seeking to enhance digital security, information integrity, and system reliability in AI-driven operational environments