The Basics of Artificial Intelligence

Armaan Merchant
Nov 5, 2018 · 7 min read

When you think of Artificial Intelligence, your thought immediately goes to robots or anything automated. Even though you’re right, what if I told you Artificial Intelligence is so much more, than just robots.

Top Google Assistant. Bottom Siri.

Take a look at the image to the left. You could probably tell what they are. The top one, is Google Assistant, found in Google’s Pixel Phone line and the bottom is Siri, Apple’s friendly assistant found on virtually every Apple device. Both of these virtual assistants are examples of Artificial Intelligence.

Think of it this way, robots are like shells and the computer is the actual AI. It’s like the human body. Your brain controls your movements exactly to your specifications. This is the same concept used in AI. The computer resembles the brain and it controls all of the movements the robot or machine makes.

AI can manifest itself in many different ways, to name a few we could have machines in an assembly line, all programmed to do a specific task, we could have robot vacuums that vacuum our house whenever we want or we could even have AI taking customers’ orders at restaurants. These tasks are made to make human life easier.

So, the question arises, what is Artificial Intelligence?

Artificial Intelligence is an umbrella term, covering two main subtopics, Machine Learning, and Deep Learning. Artificial Intelligence was designed so, you don’t realize that a machine is doing the job, exactly how a human would.

AI is basically a broad area of computer science, that allows machines to complete similar tasks to humans. For example, AI could easily be used when programming an autonomous vehicle. You could program it to stop at red lights, proceed with caution when passing someone biking or even program it to be able to sense pedestrians jaywalking but, the point of AI is to exhibit human actions. So, if there is someone jaywalking, you could teach the computer to exhibit road rage, like honking the horn and yelling at the jaywalker to move out of the way.

Many of you may not know this but, Artificial Intelligence is not a new concept. It was actually coined in 1956 by Dartmouth Professor, John McCarthy. As a child, McCarthy was fascinated with any mechanical machine out there, like automobiles or planes. His curiosity sparked when he wanted to know if the machines could learn exactly as children do. He used trial and error to develop formal reasoning, which is how you can determine weak and strong AI.

John McCarthy’s theory was proven when computer scientists realized they could actually, teach the AI form (robots, computers, etc.) to do what they needed. It’s how we learn as well. We learn by reading and scanning textbooks while, the computer learns the same way, by reading and scanning the internet. All of this learning is done by Machine Learning.

Machine Learning

Machine Learning is a type of AI that teaches the machine to act on its own. Similar to humans, machines can retain information and get smarter overtime. However, unlike humans, machines aren’t susceptible to things like short-term memory loss, information overload, fatigue or distractions. You could say, it’s a software that has the ability to think on its own and perform tasks without being explicitly told to.

After learning all this you may be wondering, how do these machines actually learn? The machines scan through billions of photos a minute to be able to identify anything you want it to. You just need to program what specifics the machine needs to look for.

If I wanted to teach my machine to identify, a cat or a dog. I would want to make sure that the properties of those animals are what the computer looks for when diagnosing the animal.

I could tell the computer to look for specific things about the appearance of the animal like:

  • A cat has very prominent and noticeable whiskers while a dog normally doesn’t.
  • A dog has a short tail, while a cat’s tail is very long.
  • While standing, a cats’ legs are close together while dogs’ legs are further apart.
Image result for machine learning diagnosing a cat from a dog

They analyze all this information using something called neural networks. Neural networks are basically like a filter for the AI to process and analyze all the information they were looking for. They are able to filter out write and wrong information using linear algebra.

Weak AI

Like humans, Artificial Intelligence learns the way we do. We are given a textbook or a website, and we retain the information we’ve read. AI does the same thing, but in a more efficient way, as it has the whole Internet at its disposal.

Fun Fact: Did you know that 90% of the world’s data, was created in the past two years! That’s crazy!

In simple terms, weak AI is Artificial Intelligence created to do work on a specific task. It has a limited or “weak” amount of knowledge.

For example, assume the robot to the left was programmed to only tell me what the weather is. If I ask it what is the weather, it would be able to tell me, “three degrees celsius, with a high of seven degrees celsius”. This is because all of its information is coming from one source. However, if I ask it to tell me the time in San Francisco, it’s going to get confused and not be able to answer my question. This is because I haven’t programmed it analyze a world clock.

Weak AI is simulating thinking and does not have the ability to think or make decisions on its own.

Strong AI

Machine Learning is the key component in developing strong AI. Essentially, the AI form has mental abilities and functions that mimic the human brain. In the philosophy of strong AI, there is no difference between the form of AI exactly imitating the actions of the human brain, and the actions of a human being. Developing its power of understanding and its consciousness are still trying to be developed and implemented.

Google created an AI bot of their own. This bot is considered strong AI. It has the ability to come to conclusions about itself and is aware of its presence on earth. Google’s Artificial Intelligence Bot even stated that “the purpose of living is to live forever”. This shows that the bot is even aware that it doesn’t possess human parts, and is, therefore, “immortal”.

Fun fact: John Searle, the American philosopher that coined the term strong and weak AI, agrees with the notion that machines can have such consciousness and understanding because he wrote, “we (as in humans) are able to be such machines as our minds can mimic the behaviour that any strong AI can.” He is one of the only philosophers who believe in this AI phenomenon.

The concept of strong AI is very philosophical, as humans can’t define what intelligence really is. It is very difficult to give clear criteria as to what would count as a success in the development of strong AI. This is because it is a different perception of AI, which makes AI equal to the human mind. It stipulates that a computer can be programmed to actually be a human mind, to have beliefs, perceptions and other cognitive states that are traditionally, ascribed to humans.

Watson is IBM’s Supercomputer

An example of strong AI is IBM’s Watson. A supercomputer that has been taught, literally, everything. In the photo above, it is playing Jeopardy with the two smartest players to be featured on that show, Ken Jennings and Brad Rutter. Watson clearly won because of his ability to scan the internet at rapid speeds (to be able to correctly respond).

The way I see it, strong AI will be the future of this society. As time goes on, I believe that humans will be able answer all the questions we have about philosophy and life.

Key Takeaways

  • Artificial Intelligence is a broad area of computer science that can manifest itself in many different capacities.
  • Machine Learning is a type of AI that can teach forms of AI, without being explicitly programmed to do.
  • Machine Learning uses neural networks (and linear algebra), to learn content and answer any questions someone may ask it.
  • Weak AI is created to do one specific task, and that task only. It cannot do more than what it is programmed to do.
  • Strong AI is AI that has the ability to think on its own, like humans. Strong AI uses Machine Learning to learn about everything.
  • AI will be the future of our generation, so be prepared!

If you enjoyed this article, please follow me and clap this post below!

Data Driven Investor

from confusion to clarity not insanity

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store