What an odd way to phrase that question. It's almost like you expect to put numbers into the computer and it magically transforms into a gundam or something. Obviously, AI is useful, and we're better off knowing it than not knowing it, much like how owning guns doesn't cause ZOG to effervesce into the ether, but is obviously better than not owning guns. Back in November, somebody took it upon himself to do the math exposing electoral fraud. He composed the information into infographics, and put the results on a website. I was kind of shocked. What an eminently *useful* thing to do, any portion of which I (and I suspect any of us here who aren't too busy doomerposting) could have done, and it hadn't even occurred to me to do any individual part of it. Obviously, people with the know-how should be using it for things like this. Now imagine, you have access to tremendous firehoses of data, ranging from shit being posted on twitter, to the language being used to news articles, to government proceedings. I don't know what to do with it, but I'm not going to act like some sort of gap-toothed nigger and say there's nothing to be done. Machine learning can be a component in trading systems. Everyone's a bitcoin genius at the moment, but imagine having reasonably consistent passive income because you have a program that can trade semi-autonomously. Such applications are infinite in number. I'm also tired of hearing people (mostly wignat retards) talking about how "AI is inherently based" like it's an immutable law of nature that any machine learning system exposed to unfiltered data will become racist and constantly confuse africans with gorillas. It isn't. They would do well to understand why. I mean, this shit is literally the future, and with things like Tensorflow out there, any mentally-ill tranny can learn to us it. But not us RACIST UBERMENSCH! We mean to walk into the future as blind and technically inept as grandmas, because we can't conceive of middle steps toward any kind of victory.