> It’s extremely challenging to predict what’s going to happen even at a high level ... Central planners or regulators who seek to intervene in the market canno

> It’s extremely challenging to predict what’s going to happen even at a high level ... Central planners or regulators who seek to intervene in the market cannot hope to aggregate enough information I can get a text message from my bank within minutes of making a purchase on my debit card at the drugstore. And I know that drugstore's inventory system updates in real time too. Meanwhile, Amazon is building grocery stores that eliminate the need for checkout because their inventory updates at the very moment the consumer picks up the item. It looks to me like we have all the technology we need to gather and aggregate mind-boggling quantities of information, so unless you have more detail about those challenges, I find the claim they're insurmountable to have the same weight as the claim last century that people could never walk on the moon. You do also say that the information is inscrutable. I take that to mean that even if we have the information, we either can't process it fast enough or we don't know how to process it at all. But I'm looking at the history of science and the pace of progress in technology--both theory and practice--and I have to wonder how long that's going to be true. Or, put another way, why should we expect it to be fruitless to explore the application of machine learning to an entire economy? Why does it need to be something sacred that we never question and never dare to answer?