Much of the larger and more profound impact that AI will have on our world is in the distant future. But we are seeing an incredible array of advances in technology occurring now—some of which will prove useful, and others, perhaps, not so much. Let's touch on a few of the more promising ones here.
One of the more interesting advances in recent years has been the introduction of generative adversarial networks. The principle behind this technology is fairly straightforward: two algorithms are developed.
The first algorithm is designed to generate a thing—this can be anything, such as a picture of a cat. Obviously, without having first been taught how to generate a picture of a cat the algorithm just generates some random image that it presents as a cat.
The second algorithm has been trained to recognize that thing—such as a picture of a cat in this example.
The two algorithms are then hooked together, with the first algorithm generating pictures...