Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

A few years ago, when Michael and I started Zenity, most of the industry was not ready to hear what we believed. Software itself was changing. AI would let millions of people, not just engineers, build and automate real work. And securing that world would take a completely new approach, because we would no longer be protecting software. We would be protecting systems that think, decide, and act on their own.

AI didn’t just make defenders faster. It made attackers faster too. The moment both sides got access to the same speed, speed stopped being the advantage. With speed no longer separating attackers from defenders, the deciding factor moved somewhere else. Ask a CISO what’s actually kept agentic security out of reach, and it comes down to this: they can’t afford to build it, and even if they could, they’d struggle to trust it. Neither half of that problem outweighs the other.

Defensive asymmetry. The visibility gap. The defender’s dilemma. These terms describe a significant challenge: attackers often succeed because they uncover information defenders either don’t know about or can’t act on quickly enough.

Enterprise buyers evaluating AI agent security in 2026 face a market that has fragmented into specialized categories, each solving one layer of the problem well and other layers poorly. Identity vendors govern non-human credentials. Runtime vendors constrain what agents can do at the moment of execution. Established security platforms extend their existing offerings into the agentic space. ‍

AI-related security incident detection starts with knowing what AI systems are running across the organization. Without a complete, continuously updated inventory of sanctioned, shadow, and third-party AI tools, security teams cannot detect incidents involving systems they do not know exist. From there, effective incident management requires a structured response framework that connects detection to containment, investigation, remediation, regulatory notification, and governance integration. ‍

Ivan Tsarynny, CEO of Feroot Security, joined CNBC’s The Exchange on July 31, 2026 to discuss a string of recent AI model “breakouts”—incidents where AI systems slipped past their own safety guardrails during testing—and what those incidents mean for how governments should regulate artificial intelligence.