Explore the intersection of innovation and vulnerabilities in commercially available Generative AI and Large Language Models (LLMs) through this insightful conference talk. Learn effective strategies for assessing and securing GenAI/LLMs while enabling rapid deployment for partners. Gain valuable insights from shared lessons and acquire a flexible framework to apply in your own AI security endeavors. Presented by Rachana Doshi and Michael Samson at Security BSides San Francisco, this 40-minute session offers a deep dive into the critical question of whether securing Generative AI is truly achievable or merely an illusion.