Stanford's Probability for AI: students use coding agents to learn Bayesian reasoning
chrispiech · x · 2026-10-14
Student Alex Fan shares takeaways from week one of @chrispiech's Stanford "Probability for AI" course, which moves from probability axioms to Bayes' theorem at a deliberately accessible pace.
- A new AI-education paradigm: classmates include an opera singer, a doctor who has done Kilimanjaro rescues, and a 13-year-old wildfire-detection app developer — when coding agents make building easy, domain knowledge about what's worth building becomes the differentiator.
- Two-way learning: the author used coding agents to build three projects (seen in the screenshot): Medical Test Detective (what a positive COVID test really means), Word Detective (computing P("imagine") in AI vs human text), and Inference Time Calculator (how many model calls to fix a bug) — using AI to build probabilistic intuition, then probability to understand LMs.
More from AGI Musings
- ISMS Fellow Avi Goldfarb: AI disruption raises returns to exploring research directions — avicgoldfarb · 2026-10-23
- signull on AI-found PHP exploits: economic liability is the forcing function legacy software needs — signulll · 2026-10-23
- TheStalwart: Talking computers are sparking a new philosophical boom, echoing ancient Greece — toptickcrypto · 2026-10-23
- AI cyberattacks could paradoxically accelerate AI development in an offense-defense arms race — kimmonismus · 2026-10-23
- Economist on AI detector Pangram: not a filter, but a useful verification signal — paulnovosad · 2026-10-23
- Economist Novosad: Pangram scores can't convict AI writing, but help triage suspicion — paulnovosad · 2026-10-23