In brief: Most teams fail not because they build badly, but because they build the wrong thing. These are the five questions every project at MAI Labs has to answer before we write a line of code — an interrogation built to kill weak ideas early, cheaply and without ego.