To do this they need to do far more than monitor and engage, they need to be able to parse meaningful data and insights from a growing sphere of what is all too often largely irrelevant data.
It competes based on its ability to parse a problem across thousands of processors, race through its massive data reserves, evaluate a near-infinite number of options, make the relevant correlations, narrow the possibilities and arrive at what it identifies as the highest-probability correct answer.