How DeepFakes could Change the Internet?

During Rakhee, many of us receive gifts. But this Rakhee some people received really unique gifts. For example, consider Umang Agarwal, who posted this video “Hello Umang! Happy Rakhee to you! Wishing you a very Happy Rakhee to you from your sister!” Umang wasn’t the only one of a kind.

In fact, the Cadbury celebration’s ad featuring Hrithik Roshan could have personally wished people a Happy Raksha Bandhan. “Hi Deepshikha! Wish you a very Happy Rakhee!” The process was simple- First, you had to scan the QR code on the box; and then enter your sister’s name and the, automatically, a video would be generated “Hi Bhavya! Wish you a very Happy Rakhee!” Which you could then share on Whatsapp.

The whole process took only about 10 minutes. Cadbury not used it only for Raksha Bandhan but for Diwali as well. featuring Shahrukh Khan where he endorsed local/small businesses in the advertisement. “This Diwali, shop from the nearby store Choice of Fashion.” “…shop from the nearby store Ajkal Fashion.” “…Royal Fashion…” “…M K Cloths…” to help India’s small businesses in their COVID recovery because they do not have access to what big businesses possess- and that is- superstar brand ambassadors “…By making India’s biggest Brand Ambassador be their brand ambassador!” But how did Cadbury do this?

They used machine learning techniques to recreate Shahrukh Khan’s face and voice By making use of the machine learning algorithm and the viewer’s pin codes, different versions of the same ad were created with local store names “This Diwali, shop from the nearby store Choice of Fashion.” “…shop from the nearby store Ajkal Fashion.” “…nearby Royal Fashions…” “…MK Cloths…” “…buy from the nearby Laxmi collections ” and then showed them to the viewers that lived nearby these stores.

This new technological development is called “Deep Fakes” And it is being used not just in marketing campaigns but also political campaigns For example- take a look at this video: “… the Citizenship (Amendment) Bill just got passed in the Rajya Sabha” “He said they will install 15 lakh CCTV cameras. Have they installed?” “They have only bluffed you in the name of promises. But this is the time for Delhi.” You might think that all these three videos are original. But in fact, only one of them was original.

The English and the Haryanvi ones have actually been produced by the deep fake technology and they were produced by the BJP before the Delhi Assembly elections in 2020 And in this post, we’ll take a closer look at how deep fake videos are produced and what impact would it make on our society. “Deepfakes have started to appear everywhere. From viral celebrity face swaps to impersonations of political leaders” “…as the technology advances, the danger is that deepfakes will be used to mislead voters” “We are entering in which our enemies can make it look like anyone is saying anything at any point in time… You see, I would never say these things! Stay woke, B.”

Must Read Articles:

The technology of Deepfakes became very popular in 2017 when a Reddit user-produced fake corn clips of famous Hollywood celebrities Deepfake videos basically use artificial intelligence to swap out the faces of people in videos. They are called deepfakes because they use deep learning technology, a branch of machine learning wherein, if a computer is fed enough data, it can produce output on its own and thereby produce such fake images and videos. Now, let us talk about what can be the impacts of deepfakes in our society and what are the places that it is being put to use. We will divide this in two sections- Politics and pornography On the political front, the technology gained attention first in 2018, when a comedian copying Barack Obama released a video on how deepfakes can be harmful.

“Moving forward, we need to be more vigilant with what we trust from the internet.” In a separate incident which you might not have heard about A deepfake video of the President of an African nation, Gabon, was released due to which the country’s military even attempted a coup ! Deepfakes were also during the Delhi assembly elections like the videos we saw. Using this technology, videos of the Delhi BJP President, Manoj Tiwari While one video had Tiwari speak in English, “He said they will install 15 lakh CCTV cameras. Have they installed?” the other was him speaking in the Hindi dialect of Haryanvi. “They have only bluffed you in the name of promises. But this is the time for Delhi.” Neelkant Bakshi, co-in charge of social media and IT for BJP Delhi, has said in an interview that Deepfake technology was used by them to scale campaign efforts The Haryanvi videos could help them target one particular audience even if the political candidate didn’t speak that language.

You may now wonder how these videos were being produced Sagar Vishnoi, the chief strategist at The Ideaz Factory, who helped BJP Delhi, claims that his company used a ‘lip-sync’ deepfake algorithm and used the speeches of Manoj Tiwari as data to produce his English and Haryanvi videos. The firm even hired a dubbing artist to provide an audio reading in Haryanvi, which was then superimposed on the video of Manoj Tewari. Neelkanth Bakshi says that these deepfakes were distributed across around 6000 WhatsApp groups reaching approximately 1.5 crore people.

Bakshi says the response of the voters to those videos has been very positive. He said that several Housewives said it was heartening to watch their leader speak their language After the “viral” response of the Haryanvi video, the party decided to go ahead with a second video of Manoj Tiwari in which he would be speaking English , targeted at “urban Delhi voters.” Maybe it is because of impacts like these that several people are further adopting deepfake videos A research by a think tank based in UK revealed that deepfake videos have almost doubled in 2019.

Now one quetions you might ask is that will the deepfakes lead to more fake news? The answer is yes! Because the level of digital literacy in India isn’t super high and when even low-tech fake videos have led to violence, Imagine how DeepFake videos can impact our society and fake news! A second impact of the deepfake technology is that it could lead to an increase in pornographic content online. A research showed that 96% of the total deepfake videos were used for pornography.

Sensity, an Amsterdam-based visual threat intelligence company, found that such videos had increased from 8,000 to 50,000 in merely 1.5 years Major adult websites have created mechanisms to block deepfake videos yet such videos can still be found on the internet LiveMint spoke to one creator of such type of porn He said that earlier, he used to make such deepfake porn videos for personal pleasure But when his name became famous on both Indian and international websites, he started doing it for fame.

He told the LiveMint journalist that doing this might be wrong according to the people, but he enjoys doing it. And while mainstream websites might be blocking such content, but there are many upcoming websites that both produce and share such content. The main reason such websites host such content is that their wish is to make money.

This is exactly what the founder of Sensity had to say He said that when people watch such videos and comment on them, those video portals would make money out of it. Very often, they charge per view or clip and sometimes even earn through ads Often, payments are made via cryptocurrency on such websites. The report of Sensity has also revealed that Now, bots have been created that can be used on messaging apps like Telegram that can create a deepfake images or videos themselves Henry Ajder, an expert on deepfakes and the lead author of the report of Sensity, said that such bot technology has made creating deepfakes very easy for the people which has led to a rise in the number of pornographic content as well Apart from these two drawbacks, there are positive aspects of deepfake technology as well.

For instance, let us talk about the film industry Film industry can use it to create digital voices and updating film footage instead of re-shooting it. Now, if you have ever watched a dubbed movie, then you might know that the experience is not quite great Deepfake technology can make voice dubbing more realistic For example – In 2019, David Beckham shot for a global malaria awareness campaign in which it appeared as if he was using many different languages and appealed to many different audiences The technology can have positive uses in the social and medical fields as well It might be controversial right now, but if a friend or relative if yours has passed away, Deepfake technology can help you have a conversation with them And as we saw in the introduction, deepfake technology can help brands in marketing campaigns While deepfakes might have more negative impacts than positive ones, If we wish to use this technology in a more positive light, then we would need more skilled people who understand this technology.

This technology right now has a bigger negative side than a positive one We need to look for solutions to paint this technology in a positive light The first is that we will have to provide a thrust to digital literacy and provide people with the skills to help them identify a deep fake video Giorgio Patrini, founder of Sensity says that If you are watching a video that’s very sensational, your first reaction should be to question the accuracy of it. It is not too difficult to spot a poor quality deep fake With the current technology, fine details like lip-syncing, skin quality, hair, are not yet perfect IIT-Ropar and Monash University in Australia have developed ‘FakeBuster’, that helps detect deepfake videos If we want to remain a step ahead, the government and private education institutes need to fund deep fake detective technologies Private companies can also play an important role in this by helping bridge the gap between the supply demand of skills.

Authors Simran Jain and Piyush Jha have written that in India there is no explicit law banning deepfakes. Amidst the current laws in force,it is illegal to produce sexually explicit material but it is unclear whether these regulations can tackle deepfakes or not Apart from that, we also need to ponder about its effects on politics For example, the state of California in 2019 passed a bill that made it illegal to circulate deepfake videos of politicians within 60 days of an election date.

The technology might not be super-advanced right now, we need to realize that it might become advanced at one point of time and this is why due importance needs to be paid to this issue Hao Li, a Deepfake expertr said that this technology is developing extremely rapidly. There could come a time where there is no way that we can actually detect deepfakes from the originals And this is why we need to think about this issue today itself.

Leave a Comment