Testing and Red-Teaming AI Systems
Testing AI isn't like testing code. Set acceptance criteria before you test, red-team generative systems and make a release decision you can defend.
What you'll be able to do
- Set acceptance criteria and test data before testing begins
- Evaluate results by group so average performance doesn't hide harm
- Plan and run red-team exercises on generative AI systems
- Make and record a release decision, including accepted shortfalls
An AI test and red-team plan with results
What's in it. Intended use and risks, acceptance criteria set before testing, test data with group coverage, evaluation results by group, red-team scope and findings, the release decision and accepted shortfalls, and retest triggers with the regression suite.
You build it lesson by lesson, using your own organization, and submit it for your certificate. It's yours to keep and adapt.
Who it's for
ML engineers, QA leads, AI governance and security teams.
What's inside
- 1Where the requirement comes from 18 min
- 2How small teams and enterprises meet it 16 min
- 3Criteria and test data 22 min
- 4Testing generative AI 22 min
- 5Red-teaming 24 min
- 6Independence, pipelines and decisions 20 min
- 7Hands-on lab: small team choose one 22 min
- 8Hands-on lab: enterprise choose one 24 min
- 9Finish and submit your deliverable 20 min
Built for your size
The requirement is the same everywhere. How you meet it depends on who you have. You pick the lab that matches your organization.
How you earn the certificate
Submit your finished deliverable, which is scored against a published rubric, and pass a short scenario quiz. Your certificate goes to your wallet and can be checked by anyone on the public register.