What happens when we teach a computer how to learn? Technologist Jeremy Howard shares some surprising new developments in the fast-moving field of deep learning, a technique that can give computers the ability to learn Chinese, or to recognize objects in photos, or to help think through a medical diagnosis. (One deep learning tool, after watching hours of YouTube, taught itself the concept of “cats.”) Get caught up on a field that will change the way the computers around you behave … sooner than you probably think.
That is the description of this TEDx Talk on the hubris of a technologist, er, I mean, the science fiction possibilities of technology run amok.
As much as we look back and see how much technology has aided humans, there are still those who look ahead through a darkened glass. Here is yet another technologist who believes that soon we’ll all be out of work and will just be lazy, good-for-nothing slaves to technology.
Time after time we’ve seen how technology has greatly improved the lives of humans. People have been removed from dirty, dangerous, demeaning work. Yet, there remain many other places to tackle that same problem—I can think of mining for an example. How many miners do we need to lose to cave-ins or black lung?
The main problem I see with Howard’s analysis is the old “linear extrapolation” trap. He says in effect that this technology can never stop growing exponentially. That, of course, is false. Nothing in nature continues to grow forever linearly. It will either level off or be asymptotic to some axis.
On the other hand, he does have some fascinating examples of technology helping humans.
