AI Lab Workshop 4 - AI Evaluation & Testing

A free 90-minute workshop at London Business School teaching how to evaluate and test AI products for accuracy, reliability and safety.

This is the fourth session in the AI Lab workshop series, led by mentor Ivan Ante. Participants learn to design realistic user tasks and test cases, evaluate AI outputs for accuracy, relevance, consistency and safety, and identify failure modes such as hallucinations. The session shows how to use test results to prioritise improvements and guide the next product iteration. It is primarily aimed at LBS AI Lab participants but open to anyone building or testing an AI product, and is best suited to those who attended earlier sessions and have a prototype to test.