What is AI? Define Machine Learning? Is AI threat?
By reading the title you already guessed what we are going to discuss here. Yup, it’s artificial intelligence and machine learning, one of the most trending topic nowadays. You may have seen many science fiction movies and books related to these topics. But in reality, what is AI? How can we define machine learning? Is AI threat? Stay tuned and I am gonna answer all your questions.
Artificial Intelligence (AI) and Machine Learning (ML)
While scrolling through the internet and news feed of your social media, you may have heard of the terms Machine Learning and Artificial Intelligence. Talking about ML and AI, there is a thin line difference between both of them. Many times, people use them interchangeably. But still, there’s a difference which we need to understand.
And to understand this, you are in the right place. We will discuss both of them one by one. First, we will see what is ml or machine learning, then will move forward to ai. So, let’s discuss each of them in detail.
What is Machine Learning?
I will not complicate things here to define machine learning. So, in simple words, it is basically a technique to teach a machine or making a computer or machine to learn something. Yes, that’s it! But it’s not as simple as it seems.
It may be possible that you are going through the term machine learning for the first time. But just opposite to your expectation, you are using it for a long time. It’s with us from the time when computers first stepped into this world. Surprised?
Yes, from that time we are continuously teaching the machines to perform specific tasks. To get this, let us take some examples.
The most relatable example of this is the Spam box of your email section. You may have noticed that some of the mail go to your inbox while other emails land in the spam section.
The algorithm of your email provider is designed in such a way that it can predict whether the incoming mail is a spam or is a useful mail for the user. It does so by scanning a combination of some specific words in the subject line. Or when many users mark a particular email id as spam.
So, how it works?
You may have also noticed that it doesn’t work perfectly every time. Sometimes are useful mail lands into the spam section and vice versa. This is the reason when you subscribe to a platform it says to check your spam section also for the confirmation email.
Another example from which you can relate much is Google Photos. When you search ‘dog’ in Google Photos, it shows you all the pictures of dogs from your gallery. Okay, I will give you a trick here.
Whenever you are finding a picture of a document in your gallery, just search ‘documents’ in Google Photos. It will save you much time. This is machine learning. The application is taught to identify dog, car, document or even a person. Again, it is not as easy as it seems. There are very complex algorithms running behind this in the background.
Recommendations on E-Commerce
Whenever purchasing a product from Amazon or any other e-commerce website, you see some recommendations under your product. These recommendations are based on your previous purchases, previous searches or the current product.
The algorithm of these recommended products is also based on ML. Experiments have proven that showing recommendations increases the number of sales.
You may have heard a lot about these cars. They are kind of trending. These cars are continuously taught to identify obstacles, signboards, other cars, people and all other stuff which we see while driving.
If there is no machine learning then the algorithm of the car cannot recognize the upcoming speed breakers or the signboards or a speeding car. To make the driving of a driverless car like a real driver, they have experimented continuously.
Simply saying, it is all about prediction power. The more a particular machine learns, higher is the prediction power and higher is the accuracy of the results.
It is a professional term for the algorithm which can determine the output based on whatever the data is being fetched. Depending upon the data you are providing to the classifier it predicts the result as for the algorithm.
More the data you provide higher will be the capability of prediction of the machine. Thereby, generating more accurate results.
Myth and Hoax about machine learning
There is a common hoax among people that with the advancement in machine learning, especially among the programmers. It is thought that the demand for programmers will decrease due to the advancements in machine learning.
The common thinking is that if machines will learn coding and algorithms which were earlier done by humans then there will be no need of programmers. But if someone says this, then you can say that he or she is wrong. I tell you why.
According to the experts with the advancements in machine learning the requirement of programmers will increase. It will create more opportunities and recruitments for the programmers will increase. As there will be a requirement of skilled programmers who know ML.
So, I see this as a win-win situation. After discussing ML let’s see what is artificial intelligence.
What is AI?
It is a branch or field in computer science which deals with the intelligent or intellectual behavior in computers or machines. Or saying in simple words, it is the ability of a machine or a computer to imitate human behavior.
The computers can perform typical mathematical calculations in a fraction of seconds. But they are unable to make decisions as we humans do. It is unable to tell if a particular person is angry or sad or happy.
It is classified into many different categories. We are using some of the kinds of AI from the past few years and are already used to it. But some others are still needed to be developed more.
So, let’s see the different type of Artificial Intelligence. It is basically classified into 3 categories:
- Weak AI
- Strong AI
Weak Artificial Intelligence
It is also known as Artificial Narrow Intelligence. This kind of AI is specially designed to perform a specific task. It focuses on one narrow task or it can do only one task properly. We can see a lot of examples of Weak Artificial Intelligence around us in our daily life.
It is used in Air Traffic Control. The very famous personal assistant Siri in iPhone is another example of Weak AI. The example with which you can relate highly is Computer Chess. It is one of the oldest and finest examples of Weak AI. We all have played it since we were a child.
Strong Artificial Intelligence
It is also known with the term Artificial General Intelligence. When the intellectual power of the machine or computer reaches near the human mind, we can say refer to it as Strong AI.
At that stage, the machine or computer or robot would be able to perform the tasks of humans. The complications will be similar to the human mind. It is not present in current time but is estimated that we will be able to achieve this kind of AI till 2050.
It is also known as Artificial Wide Intelligence. Experts believe, at that time everything will be covered by the android robots. If you are thinking that it’s the end, let me tell you it’s not! There’s something more astonishing after this.
This is something interesting and always seems to create controversy. Let’s discuss it. Since the principle of AI also follows machine learning. So, when a machine will learn something it will never forget it.
This principle will lead to an increase in the level of AI. A group of artificial minds will learn something. Then they will stimulate together to reach to another higher level of intelligence. After reaching a higher level, they will again stimulate to reach to another level.
This process will continue until they reach an extreme limit. This extreme limit is known as Singularity. Once the level of Strong AI or Artificial General Intelligence is reached, it will improvise to the level of singularity within very less time.
The level of intelligence will increase exponentially. Singularity is the level of intelligence in front of which human intellect will be nothing. It can also be called Artificial Super Intelligence.
AI vs ML – What’s the difference?
When saying about machine learning, in this, a machine is taught to become an expert in a specific task. For example, software which is able to identify something handwritten in English will not be able to identify something written in any other language.
This is because it’s just the expert of English handwriting. Or a software which can translate one language to another cannot identify the image of a dog or cat. Machine learning is like an Ace in the deck of cards.
But when talking about artificial intelligence, it is a system which has some intellectual power just like us. It can analyze multiple patterns and process them to perform a task. Being a jack of all trades, when given a different task it will still try to perform it, unlike ML.
Machine learning, Artificial Intelligence, and data mining are like three interconnected vertices of a triangle. But somehow machine learning is closer to data mining rather than artificial intelligence.
Talking about the scope, it is limited in case of machine learning. But in the case of AI, it is too wide and there is still much to discover in it.
Artificial Intelligence – All around Us
There are a countless number of examples which I can give you on Artificial Intelligence. Here are some of them:
Another example if AI can be seen in the security sector. The AI-powered CCTV cameras alert us when there is a movement in front of it.
Now some cameras are also tested which can identify a gun. And you won’t believe, its accuracy is more than 99%. These cameras are already in use in some parts of the world.
Predicting Market Direction
Many finance-related companies also take help of AI. In the stock market also, AI is widely used. Companies put the data of previous years or months and try to predict the direction of the stock. And these analyses are done several times a day.
The face detection feature in smartphones is another great example of AI. The best part about is that it can identify you even if you have grown your beard or shave your head.
The example which I love the most of AI is Google Lens. It can identify the text, object, building or any monument in front of it.
In Movie Industry
Now AI has taken a step further. Producers and Directors are now using Artificial Intelligence to predict which parts of the movie should be put in the trailer to make the movie a blockbuster.
To do so, they are collaborating with Google. By analyzing the trailers of previous blockbuster movies, the trailer of the new movie is made. The goal is to make the audience love the trailer and earn a maximum profit after release.
This is again one of the most relatable examples I can give. Nowadays the camera of smartphones has become smarter. It automatically detects the scene in front of it and enhances the colors in the scene accordingly.
You may have also used the Portrait mode in your camera. Here the camera sees the scene in multiple layers, detects the subject and blurs the rest of the background.
CAPTCHA is an acronym for Completely Automated Public Turing test to tell Computers and Humans Apart. It appears when you fill a form or log in to some website. Yes, I know that it irritates you. It irritates me too.
This irritating use of AI is based on the Turing Test. The Turing Test was designed to distinguish humans and machines or computers. That’s why you see sometimes a sentence like ‘I am a human’ or ‘I am not a robot’ is written under the CAPTCHA.
The CAPTCHA is designed in such a way that the success rate for humans is more than 80%. But on the other hand, the success rate for computers is less than even 0.01%.
But why is it necessary? A computer won’t try to login to a website. Yes, you are right a computer won’t try to do it but there are some spammers who do. They try to create hundreds or thousands of fake ids using their computer-generated bots.
But the CAPTCHA stops them to do this. Until the bot fills the CAPTCHA it can log in to the website or the portal. So, rather being an irritative process, it is a way to enhance the security of our personal data.
If you know some other examples too or can think of some, let me know in the comments.
Human Brain vs Computers
Before starting on this topic, I need you to answer a question honestly. Which is better our mind or a computer? Tell me in the comments what you think.
If your answer is the computer, then sorry to say but you are wrong. Let me tell you why you are wrong with a proper comparison of the human brain and computer.
Our mind or brain is a huge complex network of neurons. The cells present in the brain are called neurons. The count of neurons in our brain is not confirmed but it is estimated that an average human has about 100 billion of these pretty cells.
These neurons are like very small units of our brain. Each neuron has multiple input nodes and output nodes. These nodes are connected with other neurons and create a very complex web or network in our brain.
Talking about computers, there you get the basic unit as transistors. They are very high in number. A high-end processor can have up to 10 billion transistors. But the transistor has just 0 and 1 as information while the human brain has all the emotions and thinking power.
If you say that the computer can do mathematical calculations much faster than us, then yes you are right. But see, its processing power can be higher than us but its thinking capability is almost zero.
If our seniors tell us to not to do a certain task, then we can understand it. But the computer can’t. We can understand how a person is feeling, we can connect to their emotions but a computer can’t.
We, humans, are a result of millions of years of evolution. But the computers are here from just a few decades. So, now I think you understood what I am trying to conclude here.
Is AI threat?
It is thought that once we achieve singularity, we can even perform the so-called impossible tasks with the help of the super-intelligent robots. But there are some doubts that the singularity can be a threat to humans and other species on earth.
Even the top minds on the planet like Bill Gates or Elon Musk put a doubt on it. According to some people, there’s a solution to this. It is that we can apply some lock codes in the algorithm which says that the machine or the robot should not hurt any human.
But here is another face to this which I can think about. Since the singularity is much better than humans, so, will it not be able to reverse the algorithm and break the code which was written to not hurt humans?
But here is another fine solution. It is putting up emotions in their system. Since singularity’s intellect is much higher than humans, then their emotions will also be positive. Also, they will definitely have better decision-making power. So, it will decide to not break the code of conduct made by humans.
With the advancement in technology, the scope of Artificial Intelligence and Machine Learning is continuously growing. Some experts say that AI will be a boon for us but some also put a doubt on it.
Some say that no matter how much you try to make it perfect, it’s still a machine, there are chances of error occurrence. But here I need to know your opinion. Tell me what you think about it? What’s your opinion on this matter?
What do you think will be the effect of advancements in Artificial Intelligence? Will it be a boon or a curse for humanity? Let me know in the comments section.
Artificial Intelligence and Machine Learning are one of the most trending topics now. A computer can’t share this article with its friends but you can! You know what I mean.
A big thank you from our heart for reading out this article and spending your valuable time with us. Thanks for supporting us. Your continuous love towards our content makes our efforts successful.