Artificial Intelligence (AI) still sounds like something out of a sci-fi film. A super smart robot with dreams of his own, for example. The truth is, AI is all around us, it’s already here and slowly wiggling its way into our everyday lives. Amazon’s Echo and Dot, with the AI interface we know as Alexa, understands spoken commands and performs many actions- she’s more than just a speaker for your music. The Nest thermostat gets to know your daily schedule and turns the temperatures up and down to suit your needs. Using your phone’s location services, Nest knows when you leave. It can sense when you walk into a room and light up, showing the time or the temperature. It can even send you an alert if temperatures are cold enough for your pipes to burst.
Knowing that AI is all around us is both comforting and disconcerting. More disconcerting, though, are the other buzzwords flying around. AI and machine learning have been used interchangeably up until now, but the fact is, they’re not exactly the same.
Here are 5 things to know about AI and machine learning:
- AI, or artificial intelligence, means developing computers that are able to perform tasks that (up until now) would have taken human intelligence to complete. They may require visual perception, speech-recognition, decision making, and translation between languages.
- Machine learning is the method with which we achieve AI. The idea is to create computers that take initiative to learn without being supervised. Google is an example of machine learning, as are Facebook and LinkedIn. Ever wonder how Facebook knows to recommend friends to you?
- Artificial intelligence through machine learning taught itself how to win at checkers by playing against itself thousands of times. It learned the strategy on its own and was then able to beat The Connecticut state champion. The unbelievable part- this was done in the early 60s.
- A more complex example of machine learning is a self-driving car. Engineers are tackling it in the same manner as the checkers. Letting the software ride along on hundreds of roads in all parts of the US and recording data about steering, accelerating, and stopping.
- Deep learning is a type of machine learning. It’s a relatively new area, and one that would require far more than a paragraph to do it any justice. The idea is to bring machine learning closer to its original goal of being intelligent and doing things as good as, or better than, people. It strives to allow a computer, over time, to learn how a contextual model should be structured, it will perceive the world through that model, use that model to reason and make decisions, and will someday start to take that data to the next level. For example, a Twitter-bot having conversations with more and more meaning over time as it learns from what others have posted. Creepy or cool?
There are many implications to artificial intelligence — not all of them end with us humans creating an anthropomorphic computer that transcends humanity. There are grand possibilities in the field of healthcare, cyber-security, pharmaceuticals, and now even your local IT service desk.