AI-Driven Cyberattack Risks - A Foresight Study

Abstract

The digitalization has integrated artificial intelligence (AI) across multiple domains, including cybersecurity, creating both opportunities and risks. While AI enhances organizational defense through automated threat detection and incident response, it also enables scalable and autonomous cyberattacks by threat actors. This study examines future risks posed by AI-driven cyberattacks using foresight methodologies that combine horizon scanning and scenario development to identify current and emerging offensive AI capabilities. Previous studies indicates that present AI-enabled threats primarily involve social engineering and adversarial attacks, while future risks may include quantum-powered AI and fully autonomous systems. The study highlights a widening gap between offensive and defensive AI, driven by incentives, technical skill, and expertise. Three cyberattack scenarios involving AI tools were developed and validated through interviews with experts to assess plausibility and future relevance. Finally, the study proposes precautionary actions to strengthen organizational resilience, emphasizing anticipatory governance and cross-industry collaboration to improve threat intelligence accuracy