Cybersecurity is heading into a make-or-break moment as AI reshapes how attacks are created and launched. Scott Harrell, CEO of Infoblox, shares an article on Fast Company arguing that the traditional, reactive security models simply won’t hold up in an AI-first threat environment.
Harrell points to three trends leaders need to understand now:
- AI-powered, highly personalized attacks
Attackers are using AI to study individual organizations and craft custom malware, phishing, and deepfake-driven social engineering campaigns. These attacks are often brand new and designed to evade existing tools, making “detect and respond” approaches too slow. Relying on employees as the last line of defense becomes unrealistic at this level of sophistication. - A rapidly expanding attack surface
IoT devices, aging network infrastructure, and AI systems themselves are becoming prime targets. As AI is embedded deeper into enterprise software, compromised models could behave like insider threats with broad access. Security teams will need to protect not just endpoints, but the AI and infrastructure running the business. - Cybercrime-as-a-service goes mainstream
AI is fueling an underground economy where attackers no longer need deep technical skills. Ransomware-as-a-service, exploit kits, and stolen access marketplaces make advanced attacks easier, faster, and more scalable, blurring the line between casual hackers and organized crime.
Harrell’s message is straightforward: simply adding AI to legacy security tools creates a dangerous illusion of safety. To succeed in 2026, organizations must rethink cybersecurity from the ground up—shifting from reactive defenses to preemptive, adaptive strategies built for AI-driven attackers.





















