AI & Cybersecurity: Dan Boneh Interviews Nicolas Carlini

Watch Dan Boneh, the Co‑Academic Director of the Stanford Advanced Cybersecurity Program, interview Machine Learning and AI Security Researcher, Nicolas Carlini, one of the industry experts featured in our newest course on AI Security. These two cybersecurity experts discuss the history of AI Security, adversarial training, model training, and more. [0:00] Introduction [2:35] Adversarial examples during inference and training [26:34] Prompt injection [34:04] AI Security as a Career [38:00] Model extraction attacks [49:30] Protecting the training data [57:35] AI for Code generation [1:10:40] CS education in the future