In the previous post in this series, I talked about testability, and the various aspects of it, in relation to testing a product that’s been AI-enabled in some way. In this post, I’ll focus on a specific case study and apply the thinking from the previous post to that study.