Stanford University · Course
AI Measurement Science (CS321M)
AI Measurement Science (CS321M) is a ten-week Stanford Online graduate course on frameworks and methods for evaluating, benchmarking and understanding AI systems, taught from first principles in response to what it calls a measurement crisis.
Who it suits
- Researchers and engineers who evaluate or benchmark AI systems.
- People who want rigorous alternatives to ad hoc benchmarks.
What it covers
- Validity theory applied to AI evaluation: content, criterion, construct, external and consequential validity.
- Psychometric models for AI measurement.
- The problems of benchmark saturation and inconsistent evaluation methods.
- How to understand, implement and critique state-of-the-art measurement approaches.
Common questions about AI Measurement Science (CS321M)
- How long is it?
- Stanford lists 10 weeks, for 3 academic units.
- Is it open for enrolment?
- Stanford’s page said it was not open when we read it and showed it as possibly offered in spring; check the page.
- Which certificate counts it?
- Stanford lists it toward the Artificial Intelligence Graduate Certificate.
More like this
Keep learning on Infrato
- Path: AI for data analysts
- Path: AI for software developers
- More from Stanford University: all Stanford University listings in the directory
- Explore all AI courses
Where this comes from
Written from Stanford University’s own pages, read on 2026-10-14. Content, prices and availability change, so the provider’s page is the source. We are not affiliated with Stanford University and this is not an endorsement.