What we measured, what broke, and why the app does what it does.
My own AI's answers, and my colleagues' AI documents. I built Brief so I could understand both without reading every word.
