Could a computer ever be as smart as a human? Can computers truly think? Advances in the recent decade in fields such as artificial intelligence and machine learning have certainly proven that perhaps we are moving in the direction of intelligent machines. But whether or not you believe a computer could be as intelligent or rational as a human being, there are many things in the world around us that are benefitting from the artificial intelligence revolution. This post is not an ethical/intellectual debate about whether computers could eventually become smarter than humans (I’m looking at you, Elon Musk). This post simply hopes to highlight the good/interesting uses that artificial intelligence is being put to in today’s world.
But first of all, what is artificial intelligence? We hear that world thrown around a lot. Same with “deep learning” or “neural networks” or “machine learning”. Those are subset fields of artificial intelligence. Broadly, artificial intelligence is the field of study centered around creating intelligent computers and software. The way that that intelligence is achieved is often gone about through different sub fields, some of which really only try to tackle certain problems (ex. natural language processing) versus more general intelligence (neural networks). However, artificial intelligence is a field that requires many interdisciplinary fields to solve the grand problem of giving computers intelligence. Areas such as mathematics, linguistics, psychology and computer science often converge to give rise to new approaches in artificial intelligence research. If you’re really curious about artificial intelligence, I highly recommend reading the class notes of Stanford’s artificial intelligence class, CS 221. I have had several friends take the class and say they loved the challenge and application of what they were learning. I hope to take it sometime soon as well!
So where is artificial intelligence? A recent article on Wired magazine highlighted Twitter’s application of AI to create a highly accurate (93%) not safe for work (NSFW) detector, which censors tweets with highly explicit text or photo content from general Twitter feeds. Siri and other digital assistants are more prominent examples of artificial intelligence that use natural language processing to converse with humans. Please note that these assistants don’t “know” a language or really “understand” what you’re saying – they’re simply following commands and breaking apart your words into a series of symbols that they have been programmed with (think of a dictionary) to execute programmatic requests. Language translation is another similar example of AI. Facebook uses AI to identify people in pictures. Google has self-driving cars! (It’s up to you to decide if that’s good or bad). On a smaller scale, investors may use AI systems to invest in stocks (especially if they have many stocks to manage). Chess programs (such as the infamous Deep Blue) are another great example of artificial intelligence. Any video game with some sort of “AI” character (non-player controlled character) is another basic example of AI.
If a computer is trying to do something that a human can do, it’s probably safe to classify it as a form of artificial intelligence. So, can you think of any other examples of AI that you interact with on a daily basis? The list may be longer than you think!