Webinar
As AI agent-based systems become more common across higher education—from personalized learning to autonomous research—CIOs and CISOs face urgent challenges around risk, trust, and identity governance.
In this fireside chat, Dr. Raphael Yahalom, MIT Sloan Research Affiliate, and Noga Shachar Schleyer, Director of AI Strategy & Acceleration at CyberArk, introduce a new framework for governing the identities of AI agents—bridging human, system, and machine governance within academic ecosystems.
Grounded in real-world use cases and cybersecurity research, this session offers higher ed leaders a forward-looking perspective on securing innovation, protecting institutional integrity, and maximizing the value of AI investments. Gain insight into:
Don’t miss this chance to explore how identity-first security can help academic institutions embrace AI safely and strategically.