Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

AI is changing everything—but with innovation comes new risks. In this episode of AI on the Edge, we dive deep into OWASP's Top 10 for Large Language Models with security leader Steve Wilson (Exabeam). Discover why every tech company is suddenly talking about LLM security and how you can stay ahead. Inside this episode: Why traditional security doesn’t work for AI Learn from Steve’s new book The Developer’s Playbook for LLM Security and get actionable tips to protect your AI systems.

CISOs aren’t losing sleep over zero-days. The real threat? Their own employees. From phishing to prompt injection, discover what keeps security leaders up at night.

AI agents now move data, collaborate, and make decisions at machine speed — millions of actions per second. But our entire security architecture was built for humans, not for autonomous AI. In this new Agentic World, every action is faster, every breach more invisible, and every compliance gap more dangerous. Protecto introduces Agentic Controls — intelligent, context-aware CBAC Agents that live inside AI workflows. They understand policies written in plain English, enforce zero-trust decisions before data ever leaves its boundary, and protect privacy across industries.

Trust in AI starts with transparency. In our recent conversation, “No Time to Drift,” Sebastian Goodwin, Chief Trust Officer at Autodesk, shares how his team is putting that principle into practice — by creating AI Transparency Cards. Think of them like nutrition labels for AI: clear, consistent, and designed to help customers understand what’s inside. Each one outlines what the model does, how it’s trained, safeguards in place, and more.

Your WAF is Providing a False Sense of Security Improper network configuration can completely nullify the effectiveness of your Web Application Firewall. If attackers can discover your origin server's direct IP address: They can bypass your expensive security controls entirely. Your "internal" services become externally exposed. You have a massive, unknown gap in your defenses. This animation is a clear example of why security doesn't end with buying a tool. Proper integration and a zero-trust mindset are non-negotiable.