Let's take a knee for a moment. Let's take a pause from all the coding posts, all the various pieces of brilliant code posted by my friends at Coding Blocks and let's discuss AI. If you're looking for any sort of code in this post, then look away now. You're not going to get it. Leave. Now. Or stay, if you want to just read a fun (hopefully) piece looking at AI in a holistic manner.
When I think of AI, I go back to my distant childhood memories of watching those red-eyed behemoths hulking around a post-apocalyptic planet earth. Of course, I'm referring to the Terminators here. Those terrifying epic war scenes at the beginning of both the James Cameron directed films, 'The Terminator' and 'Terminator 2: Judgement Day' left a lasting impression on me as a 6-year-old. But, rather than cowering about the thought of Elysium and hiding in oblivion, I became fascinated by these skeletal organisms. Fast-forward 18 years and this obsessive passion in robotics and machine learning resulted in an MSc in AI. But fear not! I will not participate in any attempt at making Skynet. In a world where AI is rapidly expanding, almost exponentially, it becomes clear that the technology is here to stay. Artificial Intelligence as a buzzword is being seen almost everywhere, with people and large organisations associating the term with rather trivial applications. What exactly is this 'Artificial Intelligence', and why is it so important? In this blog post I will look to answer some of these questions for those unfamiliar with this phenomenon and discuss the plethora of undoubted benefits AI can bring to the modern-day society.
When did we start thinking about AI? The initial thoughts of having artificially intelligent beings walk the earth with humans arose in ancient history. Pamela McCorduck, an American author of a number of books on the history and philosophical significance of AI famously wrote, "there was always an ancient wish to forge the Gods". This refers to the fact that AI had been rumoured for a long, long time, with these artificial beings being given knowledge by the gods, i.e. by us. A little after humans started walking the earth, the idea of having an artificial being walk alongside said humans came about. AI has been slated as a mythical event for thousands of years which is great for science-fiction, but only now has it become easier to conceptualise the idea of AI. With technology getting more and more advanced with each coming year, the mysterious myth is slowly becoming reality. Classical philosophers started discussing, and computing human logic, how we think, the processes behind human thinking, and so on. Fast-forward thousands of years and we have the culmination of this logic with the invention of the programmable digital computer in the 1940s. 'The Imitation Game' is a movie based on Alan Turing and his research group's ventures into building the code breaker machines of the second world war. Pro tip: If you haven't seen this film, watch it NOW! This machine, based on mathematical reasoning, paved the way for scientists to discuss the plausibility of building an electronic brain. This was the birth of Artificial Intelligence as we know it. Turing also created the famous Turing Test. In 1950, he designed a test which involved a human interrogating a system against a set of parameters, and if it's indistinguishable as a robot, it'll be classed as artificially intelligent.
In the last 15 or so years, there's been an exponential increase in people attempting to make that AI. A being that we're able to hold conversations with, something that can respond to us as a human would, something that can walk, something that can feel, touch, do things a human would. In short, build an artificial human. But are we there yet? Not even close, my friend. If someone asked me now, 'Hey, Sagar! What can AI do right now?', my response would be, 'Hey fella, not much.' Sure, an AI can beat the world grandmaster in a game of chess, but that's only because a computer can perform any form of logical reasoning a lot faster, and more efficiently than a human would. That doesn't necessarily make it smarter.
Consider Siri, or any other "AI" built into the radioactive devices we carry around in our pockets. What do they do? What can Siri actually do? The phone listens to what we're saying, all the time (but that's an issue for another blog post), and it picks up certain phrases that it's been programmed to pick up. It listens to its owner, who'll say "Hey Siri!", in a loud, over-enthusiastic voice - as seen in any mobile phone advert, and Siri then, will respond to a set of arguments it's been programmed to respond to, and if it doesn't know the answer, then it'll do a quick search on the internet. That's it. Is that AI? Surely not, it's just a dumb terminal that computes some simple if-else statements.
That was a rather simple illustration of what AI is right now. It's certainly NOT the only illustration of what AI is right now. AI has been used to discover cancerous nodes in human organs, helping surgeons utilise the accuracy of robotics forged with their natural intuition, organising a network of self-driving cars and computing and forecasting the international market.
My point is, there's thousands of great applications for AI, but they're not all that fancy, and they don't take the limelight as would the latest feature set of the new iPhone. Where are we? We're still in an extremely primitive place if we talk about AI. As soon as we hit that point, the singularity - where robots will be as smart as humans, if not smarter, and someone'll ask me "Hey, Sagar! What can AI do right now?", my response would be, "Have you taken the Turing Test?"