Deepfake, the Technology of Tomorrow

With Computer Technology evolving at the fast pace that it is—is it any wonder that we’d switch over to the Twilight Zone quicker than Elon Musk is sinking Twitter? The answer is – not fast enough. With simulations and artificial intelligence being on the rise, came a new form of simulation of reality – Deepfake. Which is quickly gaining popularity and already showing ways it can be harmful to our society.

What is a Deepfake? For those asking that find the concept intriguing. Deepfake is a form of artificial intelligence derived from the Machine learning technique, Deep Learning. Where through algorithms and huge amounts of datasets, one can easily create fake clips of people through face-swapping software.

Imagine you see a video of a person talking and interacting with something, only for said person to deny ever doing that. Chances are, that is a deepfake video.

How do Deepfakes work?

With the help of deep neural networks that employ autoencoders, the AI algorithm is tasked with learning several interactive clips. This helps the model to understand the behaviors and environments sensed in the video. After the learning phase, the model then maps the features of the person in the video to another.

Think of it as a form of augmented reality where the virtual world is mapped on top of the real world, making it seem that both are a part of one another.

To work out any bugs or flaws in the deepfake, another technology known as the Generative Adversarial Network (GAN) is used. That helps smooth out any inconsistencies in the video to make it seem even more real.

Why are Deepfakes so popular?

This advanced computer graphics technique quickly gained notoriety on the internet. When videos of famous celebrities and media influencers, such as fellow Facebook CEO, Mark Zuckerberg, and former US president, Barack Obama, made the round on the internet. The only problem was that the videos were fake. But they appeared real, which was enough to intrigue everyone.

Deepfakes allowed us the opportunity to venture outside the realm of possibility and create something that had never been done before. With deepfakes, we could recreate artists that had already passed in an attempt to exhume their legacy.

The biggest example is how Paul Walker, an American actor in the Fast and the Furious franchise, was ‘brought back’ to film the movie’s last scene. During the shooting of which the actor sadly passed. This was a great way of honoring him and providing a touching moment for people to revel in his legacy.

Outside of cinema, the technology created in Deepfakes was used to modify virtual smart assistants, giving them more agency. Their ability to interact and do their job was greatly improved, making them more natural.

It can also be considered for healthcare, as this technology can synthesize data that can help research diseases. Really, the sky is the limit when it comes to what we can do other than create funny videos.

Can Deepfakes harm us?

Before the technology could even be properly lauded for doing the unimaginable, it was already being used to create discord. Getting into the wrong hands, this was the perfect chance for online trolls to create content that could be harmful to others. Taking away someone’s right to privacy and creating unauthorized deepfakes that could potentially harm their reputation.

Deepfakes can become a political and social nightmare. With scandals and cybercrime on the rise, this just became another show pony for these criminals to exploit. And if that continues to happen, it could lead to even more dangerous situations on a large scale, hindering public safety.

This is why certain precautions should be taken immediately.

At MERNsol, clients work with us for our expertise, but they love us for our accessibility. We create a mix of our creativity and drive for innovation to set a path for our customer’s growth throughout all aspects of their business.

Bahria Phase 7
Khuwaja Heights 5