Description:

Please join the Schmidt Program for a conversation with Ram Shankar Siva Kumar, Head of the AI Red Team at Microsoft. This talk will provide an overview of the art and science of attacking AI systems and its societal implications. We will walk through the evolution of adversarial examples to the now famous, jailbreaks, to ask why it is so darn difficult to secure AI systems from adversaries. We will look with a critical eye at the explosion of AI Safety Institutes and their efforts to assuage this problem.

Ram Shankar Siva Kumar is a leading expert in the intersection of machine learning and security, known for founding the AI Red Team at Microsoft. This team proactively attacks AI systems to identify vulnerabilities. He co-authored the widely praised book NOT WITH A BUG, which focuses on attacking AI systems. The book is highly regarded by industry leaders and policymakers, and he is donating the royalties to Black in AI.

His work has been presented at major conferences such as RSA, BlackHat, and NeurIPS, and has been covered by media outlets like Bloomberg and Wired. He also developed the Adversarial ML Threat Matrix, a framework that outlines threats to machine learning, with his research appearing in the NSCAI’s final report to Congress and the U.S. President.

Currently, he is a Tech Policy Fellow at UC Berkeley and an affiliate at Harvard’s Berkman Klein Center, exploring AI safety and policy implications. He also serves on the Technical Advisory Board at the University of Washington.

This event is open to the Yale community. Please register in advance.