Did you know you can design your AI system to tell you how it needs to be fixed? In episode three we talk about some real world examples of false/positives, true/negatives and all the things to confuse your brain matrix. Plus, cool ways to get feedback from your users to improve your system!