Evaluation & Operations
Model Card
A short document that describes a model intended uses, training, performance, and limitations, for transparency.
A model card is a standardized summary published alongside a model that states what it is for, how it was built and evaluated, how it performs across groups, and where it should not be used. The goal is transparency and accountability, so the people deploying a model understand its limits and risks rather than treating it as a black box. Many model releases and some regulations now expect one.
Related terms
AI Governance
The policies, processes, and oversight an organization or government uses to manage AI responsibly.
AI Ethics
The study and practice of building and using AI responsibly, covering fairness, privacy, accountability, and harm.
Evaluation
Measuring how good a model actually is, from accuracy on tests to human judgments of quality and safety, often called evals.
Interpretability
Research into understanding why a model produces the outputs it does, rather than treating it as a black box.