This session explores the defensive use of AI in business security, covering common threats to AI systems and the strengths, gaps, and applications of defensive AI. It will examine how AI supports threat detection, behavioral analysis, and SOC automation, and how it enables a shift from rule-based policy development to adaptive defense. The session also looks at AI-assisted red teaming and adversarial simulations, the role of AI and ML in malware detection and network monitoring, and its applications in cloud security and identity management. Finally, it addresses the risks when defensive AI fails—including model drift, bias, blind spots, and overconfidence.