Iva Amos is an AI security strategist and independent researcher who designs and delivers hands-on cybersecurity and AI security programs for CISO and practitioner audiences across federal, enterprise, and higher education sectors. She oversees a portfolio of 600+ cybersecurity labs serving 10,000+ security professionals annually leading a multimillion-dollar CISA federal contract. A recognized voice on agentic AI risk, Iva is the author of the OWASP Top 10 for Agentic AI Applications series, the AI Rollback Architecture article, and has delivered live prompt injection demonstrations and AI security workshops to CISOs and security practitioners at Gartner, EDUCAUSE, NICE, ISACA, and HackSpaceCon. Her research centers on a single thesis: Agentic AI has two distinct attack surfaces: the reasoning layer and the execution layer. The industry is governing only one of them.