Regression test
Also: regression tests
A test you rerun after every change to make sure nothing that used to work has broken.
The fuller definition
A regression test checks whether a change breaks behavior that previously met a requirement.
- Known failure
Preserve the case
- Change
Revise system
- Retest
Confirm the repair remains
A simplified relationship, not a screenshot or a complete design.
A concrete example
After revising a prompt, rerun the example where a condition was previously omitted.
The distinction that matters
Passing old cases alone does not demonstrate generalization; also use fresh representative examples.
Related terms
Evaluation (eval)
Testing an AI tool on a set of tasks to see how well it actually does.
Training & generationTest set
Examples held back from training so you can check how the model does on things it has not seen.
Reliability & securityValidation
Checking that something is correct or in the right form before relying on it.
Where it is taught
Test an AI workflow on work you can assessSource and further reading
Anthropic · Evaluating AI agentsThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.