Agentic Browsers and AI Payments: Secure by Default, Privacy by Design… or Are They?

AI assistants that browse and shop for you. What could go wrong? This session pulls back the curtain on agentic browsers and AI-driven commerce to ask: “Secure by default and private by design: or just the opposite?” We’ll explore how emerging agentic web browsers and autonomous payment agents work and reveal the new security and privacy risks they bring to campus IT environments. Expect a technical deep dive into real incidents (like a zero-click exploit that tricked an AI browser into deleting cloud data) and attack techniques such as prompt injection and AI manipulation. More important, we’ll examine defenses and design strategies to counter these threats, from applying zero trust principles and AI-specific safeguards to new protocols (e.g., Visa’s Trusted Agent) that help verify AI agents’ identity and intent. Attendees will walk away with an understanding of how to harness these powerful AI tools safely, ensuring that higher education can innovate with agentic technology without compromising security or privacy. If you’re curious about the front lines of AI-powered threats and how to build advanced defenses against them, this session will provide actionable insights and a peek into the future of secure AI-enabled browsing and commerce.