Overview
Explore a cutting-edge solution to the challenges of intellectual property protection and user privacy in Large Language Models (LLMs) in this 44-minute conference talk from RSA Conference. Delve into a hybrid approach that leverages Fully Homomorphic Encryption to simultaneously protect model owners' assets and ensure user data privacy. Join presenters Benoit Chevallier-Mames, VP of Cloud and Machine Learning at Zama, and Jordan Frery, Research Scientist at Zama, as they demonstrate this innovative method live, showcasing its practical applications and efficiency in addressing the growing concerns surrounding LLM deployment and usage.
Syllabus
IP Protection and Privacy in LLM: Leveraging Fully Homomorphic Encryption
Taught by
RSA Conference