The Safe Protocol: Why 67% of Employees Ignore the Rules

Why do 67% of employees knowingly make security mistakes—even when they’ve passed every training module? In this episode of The Safe Protocol on The Intersection Talks, Avinash dives into KnowBe4’s global report, "From Agentic Risk to Human Wins: Building a Culture of Security in the Era of Agentic AI." We break down why the standard model of security investment—focusing solely on knowledge transfer—is fundamentally broken. As Agentic AI accelerates the volume, velocity, and cognitive load of everyday decision-making, the human brain runs on cognitive depletion. We explore the psychological mechanisms behind the "knowing-doing gap" (like ego depletion and the hot-cold empathy gap), why "Shadow AI" is a symptom of poor workflow design, and four actionable strategies to design conditions where safe actions are the default choice. 📌 Subscribe for more insights on Cybersecurity, AI, and Psychology: • YouTube:    / @theintersectiontalks   • Written Deep Dives (Wednesdays): https://synthesection.substack.com #Cybersecurity #AgenticAI #HumanRisk #ShadowAI #BehavioralPsychology #TheSafeProtocol #TheIntersectionTalks #SecurityCulture #InformationSecurity #CISO Highlights & Timestamps 0:00 – The Flaw in Security Training: The 67% Reality Check 2:15 – Unpacking the Data: Shadow AI, Deepfakes, and Confidence Mismatches 5:10 – How Agentic AI Accelerates Cognitive Load and Decision Velocity 7:30 – Behavioral Psychology: Ego Depletion and the Knowing-Doing Gap 9:45 – The Convergence: Why Attackers Target Depleted Judgment 11:50 – Reframing Security Culture: Workflow Design vs. Knowledge Transfer 13:40 – 4 Actionable Strategies for Human Risk Governance 15:50 – The Core Question Every Security Program Must Answer