AI weaponization accelerates cyberattack velocity and exploits

The cybersecurity landscape is undergoing a fundamental shift due to the weaponization of artificial intelligence. According to the Microsoft 2026 Digital Defense Report, generative AI and Large Language Models (LLMs) are accelerating offensive operations, compressing the average exploit window from 4.2 days to just 8.3 hours. Attackers are utilizing AI for automated zero-day discovery, polymorphic malware generation, and highly context-aware phishing campaigns that have increased successful click-through rates by approximately 45%.

This rapid evolution has created an ‘AI enforcement gap,’ as noted in Delinea’s Identity Security Report. While most organizations have established AI policies, there is a significant struggle to enforce them in real-time, with many employees bypassing security protocols to maintain productivity. Furthermore, the rise of autonomous AI agents introduces new risks regarding persistent credentials and lack of traceability.

In response to these growing threats, the industry is seeing both institutional and commercial shifts. In Italy, discussions at the Chamber of Deputies have highlighted the need for national resilience against agentic AI. Meanwhile, the private sector is reacting through increased investment, such as the European startup Netsec raising $10 million to provide integrated IT and cybersecurity management for mid-market companies. Experts suggest that the future of defense lies in combining AI-driven efficiency with human judgment and adopting Zero Trust architectures to mitigate high-velocity, automated intrusions.