Stanford's Fall 2026 CS 224V course covers accountable, low-hallucination AI agents

A Stanford CS 224V Agentic AI course for Fall 2026, taught by Monica Lam, centers on a semester-long team project worth 65% of the grade, according to a post by meng shao on X. The course covers retrieval-augmented generation to reduce hallucination, SMT-based accountable decision-making, and long-horizon agents.

Lectures are built around systems including STORM, SatIR, WikiChat, SPINACH and SUQL. The post also gives attendance rules, an enrollment limit, and a schedule of assignments and proposals.