Analysis Of DeepFake Video Technology From Consequentialist And Deontological Viewpoint

Introduction

Deepfake video technology has been among one of the most important issues of discussion for over a decade. As stated in Lorant’s book, “The first known attempt at trying to swap someone’s face, circa 1865, can be found in one of the iconic portraits of U. S. President Abraham Lincoln”. After Lincon’s murder a lot of politics wanted to make credit for themselves by imposing their faces on Abraham Lincon’s body. There used to be a popular belief that “Pictures always say the truth” however even before the Photoshop, the statement was put in the question. Video however remains sacred. If someone is on video then others have the perfect backup for any quote idea or action that they want to attribute to that person. For this reason, people use the videos as a reliable source. But this is all about to change thanks to the rise of DeepFakes. So what are DeepFakes?

Deepfakes are altered videos usually of a famous person produced by neural networks. It could be superimposing a face onto a body so it looks as if they are doing something that they never really did in real life. It can lead to easily taking someone’s words and altering the content in a way that makes the face movements match the new audio which has been put in. Smartphone and desktop applications like FakeApp are built based on this structure. While some people believe that DeepFake technology could result in so many negative outcomes that it must be completely prohibited from using by people, there are others who argue that there is nothing wrong with this technology and it should be regarded as a positive breakthrough. As a result, there are different line of thoughts from both consequentialist and deontological point of view. In this paper, the positive and negative aspects of DeepFake is being reviewed from consequentialist and Deontological viewpoint.

Consequentialist perspective on DeepFake Video Technology

Consequentialism simply means that the judgment of an action can be measured by the consequences of such action. DeepFake video technology from a consequentialist point of view will give an opportunity to motion picture industry to foster the development of the movies. As an example in the Star Wars: The Last Jedi movie, the crew used artificial intelligence to make a young version of Carrie Fisher character which cost likely hounded of thousand dollars with use of an insanely expensive computer. However, some researchers made the same scene using DeepFake video technology on an average computer for fun and free. This way movie producers could make great looking videos with a much less budget. It could also help producers to bring a popular but dead actor or actress back to life using his or her image in the new movie.

“The implications of AI are far-reaching, but that’s not new information for anyone. Chatbots will take advertising industry’s customer service jobs, marketing will be automated. On the flipside, it opens up new possibilities on how the advertising industry use talent. The current system requires the time-limited talent to be on set for the entire shoot. There is also the issue of the necessary talent being in a different country, or having unrealistic accommodation requirements. What if, in the future, talent licensed their image and provided visual training data to production companies to feed into their AI DeepFake system?” In such situation, a body double could be used for video shooting purpose because the double has a lot more free time than the real model, and cost less too.

Right now without the growth of DeepFake there are lots of fake news which deceive the naive people. Because nowadays society uses social media as the main source for news, people are believing whatever they see or read on media. Another positive advantage in consequentialist viewpoint is that DeepFake technology could make people aware to not trust everything they see or hear on such platforms. When people know that the video they see might not be real, then they do not believe whatever is being shown on television and social media. Unsurprisingly, there is a lot of fake adult content created from porn scenes that have celebrity faces imposed on. “Due to recent technological developments, even those with modest coding experience are now able to manipulate pornographic videos so they appear to feature non-pornographic celebrities, or even a user’s friends or acquaintances. ” This might lead to embarrassment of people, or even put their careers at risk. For example, think about how DeepFake could affect ordinary people’s lives. Since most employers use Google to search the name of their applicants, if these DeepFake videos come up on their search they would easily reject interviewees regardless of whether those videos are fake or real.

All that is required to make a DeepFake video is a few minutes of video content of a person’s face, or a larger number of photos, which are then past onto other videos with strikingly realistic results. And the internet and social media make everybody’s photo and video easily accessible. “Considering the amount of personal content uploaded by social media users, in combination with the fast improvement of AI technology, the social significance of Deepfake content is rapidly growing. ” The adult sector is one of the biggest driving forces behind the recent surge in fake videos. However, arguably the bigger danger we face is in politics. Politicians are on camera a lot often in quite a fixed position like sitting for an interview. This makes them incredibly easy subjects for DeepFake videos. So think about if anybody can make a politician say anything he or she wants. This can easily start a new war or impact the elections. For instance, black-hat hackers could make a DeepFake video of public high officials, in which the person appears to say we are breaking our diplomatic relation with another country which can have a negative effect on the whole world. In the politic world which behaviors play a critical role, such videos might create controversy at a large scale. The opposite can be true too. In a broader sense if these videos become commonplace, perhaps it could give some grounds for politicians to deny something they actually said. Assume a celebrity or politician say something in front of camera which can make chaos but after a few days they can deny what they have said. Now the person could easily get away with his or her actions, because there is no way to prove that he or she is lying.

Then it becomes a question of how to even trust any evidence given in the legal system. Because the videos are being used in the court as an evidence, it could change whether the court rules in favor of or against people based on these videos unfairly. “After seeing so many videos made with DeepFake technology, people might be suspicious of everything that they see because they will not be able to trust information anymore. For example, what if someone calls you, claiming she is your wife, and ask for your bank password. ” How should we know that she is really our friend or our wife? Such process could put even smallest bits of humans' lives into question because it undermines the very definition of trust. If we become content with banning DeepFake videos now, it could set dangerous precedence of banning other technologies too. For example, any powerful person or organization who is in charge of controlling things could simply decide what technologies should be in use and what technologies should not. Then, we might get to a point that a lot of advancement in technology is stops because they are being banned for personal reasons.

Deontological Perspective on DeepFake Video Technology

Deontological reasoning also known as duty ethics means taking an action because you ought to, not necessarily because of the consequences, unlike Consequentialist reasoning. In moral philosophy, deontology is seen more like the study of duties according to some moral rules irrespective of what the consequences of these duties might be. We could also take a look from deontological point of view that why there is nothing wrong with DeepFake video. We have a responsibility to move technology forward without paying attention to the possible consequences that it could bring. Think about what if humanity denies using modern technologies such as internet or even cars just because of their bad consequences. Is it convincing to ban internet access because someone could abuse it? It is better to focus on intensifying the culture of using such technologies or using AI to fight against AI to detect the DeepFake videos or people should come to terms with these issues such they did before by using Facebook or Twitter fake news.

In Deontological point of view, law can be one important feature which everybody should obey. Of course for now there is not a certain rule for non-pornographic DeepFake videos, but since a lot of people have access to DeepFake technologies recently, this in turn has a great effect on how DeepFake should be managed. DeepFake pornography is a growing form of online abuse which began with images of celebrity’s faces being digitally manipulated into explicit stills and videos. However, it has become increasingly easy for anybody to be targeted, with emerging of a number of apps specifically designed for that purpose. “In May (2018), Davide Buccheri, a worker in the City of London, was jailed for 16 weeks for harassment after producing faked pornographic images in an unsuccessful attempt to discredit a female colleague. ” According to Sen. Marco Rubio, if law brings some new limits on using these technologies it could easily prevent mishandling of DeepFake video technology. There is also another important distinction that has to be considered. According to Zucconi, Permission to create a face-swap video, is not the same as permission to re-distribute it or publish it online. While a friend might accept to be face-swapped for a birthday video, he might be unwilling to share that with the rest of the world. So using such technologies may put humanity at risk because it could definitely harm individuals. “Publishing a video without someone’s explicit consent can expose them to embarrassment, bullying and online harassment. Even if you do not have malicious intentions, releasing a face-swap video could be very distressing for the person portrayed. ”

Conclusion

The purpose of the current study was to examine the different perspectives of DeepFake. The findings suggest that it is a controversial and much disputed subject. In general, it could easily be inferred that the principle of DeepFake barely addresses issues and advantages. It remains to be seen how this new technology might change the world in broad ways.