Data Science Overhype

Monday February 19, 2018

I’ve been concerned for some time that we are deep in an overhype cycle with data science. It keeps appearing on hottest job lists and everyone seems to think data science can solve all their problems.

But technologists have seen this pattern before, especially with intelligent tools. And we need to be careful about making sure we do not overpromise and we explain what the limits of data science are. It is, just, applied statistics. So we have all those limits on us.

I have an essay in the current issue of Industrial Management, the house journal of the Institute of Industrial and Systems Engineers, discussing these limits and warning readers of when they can hit them.

There doesn’t seem to be an open copy of the essay online. I will post it if it pops up. Since the printed copy just arrived in my mailbox, I would think online is not far behind.