3 min read

Imagine for a second that the year is 20XX. The remnants of humanity eke out a limited existence in the shadow of the ruins of Neo New London. For years the new ruling class, the machines, have ruled over the planet with a cold mechanical grip; hunting stragglers with ruthless tactics and superior Go playing strategies.

Sounds horrible doesn’t it? Even worse, according to most news articles you might have read over the last few weeks we’re just a few years off this grimdark reality.

Just two weeks ago Google’s AlphaGo won 4 out of 5 games against Lee Sedol, one of the greatest Go players currently in the world (Sedol won the 4th match). It’s an interesting look at how far we’ve come in the world of AI as well as how far we’ve still got to go – the turning point in the 4th match, for instance, was when AlphaGo made a mistake and did not realize it for 10 whole turns. But most of the reports that talked about the event didn’t really touch on the gravitas of the match for the world of AI and how far we’ve come without realizing it, mostly focusing on the nightmare of a robot ruled future.

What was truly special about this match? We had previously thought we were 5 years away from this actually happening. AI has had a long history of hype and disappointment. Alan Turing predicted that machines would be able to successfully imitate humans by the year 2000, and 16 years later than that deadline we’ve finally gotten them to master on of the most complex games ever created. It shows just how skill Google’s team is at solving such a complex problem, and shows us a look at the practical use AI can have in the immediate future.

When it comes to AI we’ve either always thought too big about what it can do or too small but in actuality the reality, especially right now, is somewhat in-between. Currently one of the best uses of AI and deep learning is with Google’s reverse image search function – finding sources based on matching key areas of an image. It’s a small feature in the grand scheme of things, but shows how far reaching AI can be when applied to old problems. The implications AI will have on products like voice assistants, search engines, chat bots, 3D scanners, language translators, automobiles, drones, medical imaging systems, and more is going to be huge. Sure, we’ll eventually get around to mass producing robots that can accurately adapt to the job they’re given but that’s a long way away at this stage.

Right now the focus on AI should not be on reaching the pop culture future we’ve dreamed of for years but on specific applications that make our lives easier. The more data that is collected by our current AI means the better products we can create. This will cause a circular effect where more users jump on these better products, in turn giving our AI more data to work with to create EVEN BETTER products for even more users to jump on.

So all in all, the immediate future of AI is going to be much smaller and less deadly than killer machines and human batteries. And even if I’m wrong and the worst happens we can at least have peace of mind that the Great Robot War will be pretty cool.

LEAVE A REPLY

Please enter your comment!
Please enter your name here