Skip to main content

Deepfake: The Technology of Imitation Explained

  • Author:
  • Updated date:

I've always been enthusiastic about digital privacy and intrigued to learn more about data breaches and such topics.


In this digital era, we can just stand still fascinated by how far humanity has come with the help of science and technology. Every day someone from somewhere in the world adds to the scientific achievements of humanity.

Being an active user of the Internet, you might have stumbled across an image or a video that you believed was authentic but later found out otherwise. Or maybe you never found it out to be fake. You might have been unintentionally living your life with some lies you found on the internet.

What this is all about and how to protect yourself from this, this article will cover all of it.

What is Deepfake Technology?

'Deepfake' is the term given to synthetically edited images or videos with the help of artificial intelligence. The term was coined by combining deep learning and fake.

Deepfake technology makes someone appear to be doing something or saying something that in reality they have not. It started as a tool for pornographic content adding famous celebrities' faces in place of the original person and it’s still used mostly for the same purpose. However, since the past few years, politicians have also fallen prey to this technology.

Videos of several politicians like Barack Obama, Donald Trump have gone viral saying something controversial which in reality, they have not.

How Does Deepfake Technology Work?

The deepfake technology uses machine learning to create synthetic images, audio, or videos. The main algorithm used in creating and improving the accuracy of deepfakes is Generative Adversarial Network (GAN).

There are two main networks in GAN namely Generator and Discriminator.

A generator takes a random input and produces the result based on that input. The discriminator, on the other hand, judges the output produced by the generator by comparing it to the original data that was the source for the input. Then it gives a score to it based on how close it looks to the real source.

Going through this algorithm again and again the machine learns consistently and the result gets improved.

The goal is to create synthetic images, audio, or videos that appear to be authentic and accurate. And the key to do that is to let the computer observe the source to be used as the input for a long period of time so that it can take in the tiniest bits of details of the source and the output can be made as close to it as possible.

How Generative Adversarial Network works

How Generative Adversarial Network works

What are its Potential Uses?

Deepfake technology has its uses that can be beneficial to a great extent if used appropriately. Such as,

  • It can be used just for the sake of fun unless it breaches someone’s privacy and uses it for inappropriate things.
  • It can be used to make people who are no longer in this world appear to be doing something or saying something which would fill their loved ones with joy and emotions.
  • It can also be used to dub movies, series, and such in different languages where the actors can be seen speaking the respective language themselves.

A lot of this will be in the future, but that future is not far. The deepfake technology has not yet reached the levels to actually dub a full movie or such and it's still not that hard to detect a deepfake, but that will not be the case in the near future. If used for the betterment of humanity, deepfakes can come out to be really useful.

And just like any other technology, it has potential misuses as well.

  • Deepfakes have been used and are still being greatly used to create synthetic pornography using people’s faces and/or bodies without their consent.
  • Deepfake technology can also be used to create fake political or controversial speeches by famous and influential political figures or any public figure. This can lead to riots or attacks on the individual whose deepfake was made intentionally to create a ruckus.

Can a Deepfake be Detected?

Yes, they certainly can be detected. However, detecting one on your own would require high-tech skills which everyone does not have. Not to mention that every day this technology is improving and it's becoming harder and harder to detect it. But there are certain tools and plug-ins available on the internet such as video authenticator by Microsoft that you can use to detect a deepfake.

Is Deepfake Technology Something to be Worried About?

As technology moves forward, the machine learns and improvises in creating real-like synthetic images or videos. While the technology itself poses no harm to anyone, it’s the potential misuse of it that we should be cautious about.

The most important thing that comes as a reminder with this technology is not to believe everything you see on the internet immediately. Let the government and the pros detect the media to find out its authenticity and wait before jumping to conclusions. It’s important to spread this awareness to everyone.

© 2022 Shivam

Related Articles