Model Evaluation
Model evaluation is the systematic testing of an AI model's performance, safety, fairness and robustness before and after it is deployed.
What is Model Evaluation?
Evaluation goes beyond accuracy to include bias across groups, behaviour under attack, dangerous capabilities and real-world impact. Regulators, including under the EU AI Act for systemic-risk models, increasingly require documented evaluations.
How it works
Evaluators combine benchmarks, held-out test sets, red teaming, human review and field monitoring.
← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions