Hilton San Francisco Union Square | June 5 - 6

< Back

Securing AI Model Weights - Confidential Computing and Beyond

June 6, 10:15 AM - 10:40 AM
Grand Ballroom Salon B

Frontier AI systems are becoming more capable rapidly. AI model weights are a critical component to secure - they are the culmination of significant compute, training data (trillions of tokens), and algorithmic insights and optimizations. Their security is already a matter of increasing commercial interest, but depending on their future performance on hard-to-predict tasks (such as assistance in the development of bioweapons) their security could suddenly become a matter of national security. In this talk, we'll discuss potential future security needs for frontier AI, what labs can do today to prepare, and how confidential computing fits into the picture.

About the speaker

Sella Nevo

Sella Nevo

Director, Meselson Center, Senior Information Scientist, RAND

Sella Nevo, the director of the RAND Corporation's Meselson Center, is dedicated to reducing risks from biological threats and emerging technologies. He is the lead author of an upcoming RAND report on securing frontier AI model weights from theft. He previously led the Karmel group at Google Research, which included the Google Flood Forecasting Initiative, providing flood forecasts and warnings globally alongside several other humanitarian and environmental efforts. He is a Venture Partner at Firstime, a venture capital firm that invests in climate startups. He also sits on the advisory board of ALLFED - the Alliance to Feed the Earth. He has previously been featured on Fortune's Change The World list and TheMarker's 40 Under 40, published in Nature, and has been an invited speaker at the United Nations, the World Bank, Harvard, Oxford, Yale, NeurIPS, and ICML, among other venues.