Rethinking Prevention And Recovery With Barracuda XDR

Can designing for human error become the strongest cybersecurity strategy in an AI-accelerated world?

In this episode, I sit down with Yaz Bekkar, Principal Consulting Architect for Barracuda XDR and a member of the company’s Office of the CTO, to explore why the speed introduced by AI is changing the risk equation for every organization. As automation allows teams to move faster, it also means small mistakes can scale at machine speed. Yaz argues that resilience in 2026 is no longer about trying to prevent every incident. It is about anticipating failure, containing the blast radius, and recovering quickly without bringing the business to a standstill.

Our conversation challenges one of the most persistent narratives in security, the idea that people are the weakest link. Yaz explains why safeguarding the workforce begins with reshaping the environment they operate in. When the secure option is also the easiest and fastest path, risky shortcuts begin to disappear. From secure defaults and least-privilege access to paved-road workflows for administrators, he shares practical examples of how organizations can reduce complexity, limit exposure, and support better decisions under pressure.

We also tackle the limits of annual compliance training and the cultural shift required to build real cyber resilience. Yaz makes the case for continuous, bite-sized practice embedded into everyday work, from three-minute phishing simulations that teach without blame to short, hands-on misconfiguration drills for technical teams. The result is stronger habits, faster response times, and a security posture designed for real human behavior rather than ideal conditions.

If AI is accelerating both innovation and risk, how do leaders move from a prevention-only mindset to resilient operations that protect business continuity when controls fail? And what would change in your organization if every system was designed with the assumption that someone, somewhere, will eventually make a mistake?

Useful Links