Recently, Hollywood megastar Cruise went viral on TikTok for his short videos showcasing his typical eloquence. But he has not even created an account on the short video-sharing platform. It is rather his lookalike Chris Ume creating deepfake videos of Tom Cruise just for amusement.
But Deepfake videos can be far more than a pleasing time pass and tech wizardry. It can lead to mass misinformation and even jeopardize the public image of a person or an institution or even threaten a person’s well-being and mental peace. The so-called technical mastery of video editing is seen as an evil ground posing a sinister threat to people’s lives as it’s sweeping the internet lately.
There are different techniques film-makers exploit to render environment as per their will: CGI and VFX have been the most instrumental to teleport actors from the green screen to any surroundings they choose but deepfakes takes the game to a whole new level and the concern is; you do not have to be a video-editing mastermind to execute it.
What is Deepfake?
Deepfake is a so-called term referring to image manipulating technology in which an incredibly convincing resemblance of a target person is created including facial expressions, and even voice using another person’s body.
By just swapping images, a person can be doctored into another video and show him doing the acts he or she never even participated in. This image manipulation has exploded all over the internet in recent years and drew widespread attention to its sinister capabilities.
How Are They Created?
The deepfake technology is about machine learning. A creator would first have a neural network train multiple hours on the original video with the original person to pass on AI “realization” to it. It immaculately takes the impression of the original person’s movement of talking, walking, and all the facial expression from different angles under different lighting scenarios. Then the creator superimposes the target figure’s face or the whole body onto another person’s.
There has been an ongoing debate on the internet that deep-learning algorithms called Generative Adversarial Networks (GANs) would proliferate the pitch-perfect deep fake videos in the future. It is assumed that the GAN-inspired deepfakes will be even harder to decipher than the current ones. It can only alarm us about its wrongful potential if ever becomes a reality, but at the pace the interest it has garnered, it seems inevitable that either this or another alternative will be mainstream sooner or later.
Deepfakes: An Agent of Social Chaos
In late Mach of 2019, a video surfaced showing Nancy Pelosi slurring her words as if she was under the influence of alcohol or even sick. Even worse, the then US president was so convinced that he shared the video which incurred criticism from all around except his own fan base.
Critics blamed Donal Trump and rightly so, for spreading a fictional video merely to denounce and embarrass the opposite party member who had been vocal about Donald Trump’s policies. The incident is an example of how deepfakes could spearhead as an agent of chaos. As technology becomes more conspicuous, it will reach into the hands of more people, and we can only expect worse scenarios.
The video emerged to be doctored after several testings. It was not even a deepfake video but clever editing by a pro-level individual or a group. However, it ruffled feathers, the issue was stirred, and peace of mind compromised.
It just proved how devastating the effects could deepfakes cause on the personal lives of an individual and at what scale. From politics to sports, and entertainment, deepfake videos have drawn notorious attention these days.
Facebook’s Founder Mark Zuckerberg’s found himself in the hot water after a deepfake video surfaced showing him bragging about possessing the user data of billions of Facebook users. Deepfake videos are the troubles and an answer must be developed to tackle them very soon.
Celebrities have also fell prey to deepfake technology. Many Hollywood celebrities such as Emma Watson, Natalie Portman, and Gal Gadot have fallen victims to the deepfakes. These actresses’ images have been composited into another’s (the original) bodies giving an impression that they were in the act. it is an abhorrent invasion into someone’s personal space. These fictional depictions of people in porn footage can hurt the dignity and morale and break them emotionally.
Deepfakes are the anti-culture that contrasts with the morals and ethics of creativity. It is the very opposite of what we call progress or evolution. While justifications can be made on the ground that technology’s ever-growing nature, demarcations have to be drawn.
The Unflattering Role of Social Media
We live in the social media age and any sensitive news spreads like wildfire or metaphorically speaking at the speed of light. Where people have misguided aspirations to go viral on social media, under circumstances it could be the worst-case scenario for them with a manipulated deepfake video.
Moreover, most of the deepfake videos do not go viral on the web alone, to speak in the traditional sense. They arrive on social media platforms and explodes across smartphones to people. And Proliferation if such deepfake videos could cause a meltdown in the personal lives of people.
Facebook has told that they are diligently working to ward off the threat of the deepfakes. They have even mobilized a team of expert engineers to identify the manipulated contents such as video audios or photos and remove them immediately or block their upload next time if someone tries to. Facebook also allows users to report such contests which are fictional and could cause harm to a certain individual or an organization by misleading the public.
Read this too: Facebook Hacked Again: 500 Million Users’ Data Leaked Online
Deepfake: A Bad Name to The AI
One of the key reasons deepfake is harm is that it brings disrepute to the glorious AI technology. Artificial Intelligence is the driver of future technology and its implementation will be key to next-gen 5G related to the use of broadband.
However, The AI will incur notoriety with the deepfake utilizing it for clandestine purposes. Deepfakes rely heavily on the use of AI to extract the data of one individual to merge it to another. In the absence of a powerful AI, deepfakes become inefficient. Since it is the vehicle for these fictional videos, it is a worrying aspect regarding the misuse of Artificial Intelligence.
It all comes down to ethics at last. Technology is ever-growing and that is the only thing going to happen. It never slows or halts. People with knowledge and skill will always exploit their wizardry to create something that could compromise other’s public image. But if deepfakes are done for mere sport and creativity, it can be justified too but there must be more reliable defensive tools that help to detect and confirm the video’s authenticity for correct judgment.
Check out: Ctrl Shift Face‘s deepfakes on Youtube
What is your impression of the deepfake videos? Do you think they can be limited to fun and creativity or could harbinger anarchy in the society when they go mainstream? let us know your valuable input on theis topic in the coments section below.