Zachary Robertson to Teach Stanford AI Safety Course
Stanford researcher Zachary Robertson will teach CS120 on evaluating AI safety beyond benchmarks.
TLDR
Zachary Robertson announced on X that he is teaching Stanford CS120, Introduction to AI Safety, this fall. The course focuses on determining whether AI systems are safe under deployment and optimization pressure rather than solely on benchmarks. Robertson, who works on auditing LLM evaluation and incentives, shared a diagram showing a Model to Institution arrow. The post invites readers to learn more this fall.
Combined views
—
2 Sources, first seen ago
