Making machines learn, applying machine learning to successfully solve problems, and positioning business assets to benefit from machine learning are three very

Making machines learn, applying machine learning to successfully solve problems, and positioning business assets to benefit from machine learning are three very different pieces of the "learn machine learning" puzzle. I don't know how any human would be able to go from 0 to having a good grasp of even one of these pieces in three years let alone all of them. I suppose there will be (and to some degree already are) machine learning magic wands, but they are going to be the kinds of things that suck capital out of the companies that utilize them (through loss of proprietary data and competition blocking moats.)