Introduction
AI won’t kill us, but its footprint on society is concerning.
As per AI ethics researcher Sasha Luccioni – with AI advancement, it is now high time for us to focus on the negative impacts of technology. These include AI’s current, tangible impacts, such as its environmental footprint. However, the most pressing concern is the ethical issues surrounding consent and bias, amplified using DEEPFAKE Videos and images.
Deepfake Spared no one!
From former Prime Minister Narendra Modi to POTUS Trump and Obama to Tom Cruise’s iconic stature, all have found themselves at the mercy of this advanced form of manipulation.
With many AI tools entering the market every month, deepfake technology is evolving unprecedentedly. Enhancements have broadened its capabilities beyond mere facial expression manipulation. It encompasses a spectrum of movements, including intricate 3D head positioning, lifelike eye gaze with natural blinking, and fluid head rotations, all powered by advanced generative neural networks.
If you think you are immune to this deepfake epidemic, then you are ABSOLUTELY wrong. Creating the living portrait merely necessitates a handful of authentic facial photographs. This ease of generation might raise alarms among individuals who previously believed they were immune to deepfakes due to insufficient video footage available for algorithm training.
What is a Deepfake and Why Should we be Concerned?
A deepfake refers to a variant of synthetic media in the form of video and image that has indeed taken place but has been distorted using deep learning. The term “deepfake” is a combination of “deep learning” and “fake.” Deepfakes are assembled by training deep neural networks on enormous quantities of photographs or video recordings of the individual the network is supposed to impersonate. The neural network “learns” how to chart the individual’s appearances, gestures, and other characteristics by being fed an image of the individual’s face and body actions. The model creates a novel synthesized video or photo of the same individual, showing words they have never uttered or acting or conversing in ways they have never done.
Deepfakes are concerning for several reasons:
- Disinformation and Manipulation: Deepfakes can create highly realistic but entirely fabricated videos or images of public figures, celebrities, or anyone, making them appear to say or do things they never actually did. This can be used maliciously to spread disinformation, defamation, or propaganda.
- Erosion of Trust: The increasing prevalence and sophistication of deepfakes can undermine public trust in digital media and online information. It becomes harder to distinguish between real and fake, which can have serious implications for journalism, politics, and public discourse.
- Invasion of Privacy and Consent: Deepfakes can be used to create non-consensual explicit or intimate content, which is a form of image-based sexual abuse and a violation of privacy.
- Potential for Exploitation: Deepfakes could be exploited for financial gain, such as creating fake endorsements or advertisements, or for political manipulation and cybercrime.
- Ethical and Legal Concerns: The creation and distribution of deepfakes raise complex ethical and legal questions around issues like freedom of expression, defamation, and intellectual property rights.
Further, this article will provide you with 10+ deepfake videos and ask questions about this rapidly evolving technology’s implications, ethics, and future.
10 Best Deepfake Videos that you Must Know
Deepfakes burst onto the scene in 2017 when a Redditor shared videos showing celebrities’ faces manipulated onto adult film actor’s bodies. Initially targeting public figures due to available footage, researchers highlighted the technology’s misuse potential by creating a fake video of President Obama appearing to say anything they programmed. This raised alarms about deepfakes being weaponized to undermine leaders and spread misinformation.
Notable examples include a slowed-down video of Nancy Pelosi shared by President Trump that made her seem drunk, and a deepfake video shared on Instagram depicting Mark Zuckerberg appearing to reveal sinister intentions for Meta. As deepfake technology advances, the threat of realistic-looking misinformation campaigns increases, making it harder to discern truth amid rampant “fake news” claims.
Obama Impersonation
The clip is an impersonation of the former president of the United States, Barack Obama, with a reflection on the modern technology landscape. It presents the frightening truth that these technologies, especially the deepfake capabilities, can be used to manipulate audio and video, making it seem like certain people are saying things they never did. The phrase ” We’re entering an era in which our enemies can make anyone say anything at any point in time ” describes how this technology can be a threat and instrument of misinformation and identity infringing.
Morgan Freeman Deepfake
While watching last year’s deepfake of Morgan Freeman from the Dutch YouTube Channel Diep Nep, this deepfake still seems creepy and highly feasible. For instance, the video reappeared on Twitter not too long ago, where users commented on the danger of allowing such technology to enter the political arena, like in the case of the 2024 election. People said such capacities granted users too much power since it allowed them to create new false personas to which they may frame or attribute set-ups.
President Volodymyr Zelensky
The conflict between Russia and Ukraine presents itself as an opportunity for a new sinister tactic: the use of deepfake videos to spread fake news. An example closely related to the recent situation is the impersonation of President Zelenskyy’s video, which is rendered through deepfake and shows Zelenskyy bowing his head in surrender. On the other hand, Zelenskyy posted an opposing message to confirm that his people” have no intention of surrendering to anyone.” Meanwhile, Russia faced more problems as the people clashed in grocery stores over the drop in reserves due to Western sanctions.
President Vladimir Putin
While doing the long phone-in press conference broadcast on Russian TV, Vladimir Putin denied being replaced with a body double. The President knocked out any doubt in the minds of his listeners by saying: “There is only one person capable of speaking and sounding like myself, and this person is me.” Yet, many people still point out that a body double is being actively used for some of Putin’s public appearances and while visiting the regions of Ukraine controlled by fervent defenders of the occupant.
Tom Cruise TikTok Deepfake
Beginning in early 2021, a collection of deepfake videos surfaced on a TikTok profile, depicting what appeared to be renowned actor Tom Cruise engaging in various activities, including performing a magic trick. These videos quickly gained immense popularity, amassing millions of views on the platform.
Robert Downey Jr and Tom Holland in Back to the Future
An impressive use of deepfake mechanics was demonstrated in a video of Hollywood celebrities Robert Downey Jr. and Tom Holland imposed on key scenes in the Sky Way travel film. The duplication of the performances of two familiar actors in the original film produced both curiosity and awe, combining popular authors and a favorite film of recent years. The visual component of the recording was distributed throughout social networks, drawing extensive interest from other works and arousing the hearts of audiences through nostalgia.
Better Call Trump: Money Laundering 101 [DeepFake]
It is possible to assume that this is a piece of satire or fiction. The work combines popular culture, politics, and technology. The animation, performed in a deepfake style, presents the former President of the United States, Donald Trump, in the TV show Better Call Saul – a prequel to Breaking Bad. The implicit title refers to Saul Goodman – a lawyer helping clients escape dubious crime-related situations.
Zuckerberg speaks frankly
“I wish I could…” is a thought-provoking deepfake video featuring Meta founder Mark Zuckerberg. In this video, created as part of the “Spectre” project, Zuckerberg appears to speak candidly about the concept of privacy on Meta, revealing insights that challenge the public perception of the platform.
Museum creates deepfake Salvador Dalí to greet visitors
An innovative deepfake “Dalí Lives” project by The Dalí Museum, Florida, utilizes AI to bring Salvador Dalí back to life. GS&P agency combines art and technology to create a true-to-life image of Dalí by analyzing over 6,000 video frames and applying 1,000 hours of machine learning. The visitors are greeted by an eccentric eccentric who reveals the exhibition. This blend of art and AI provides an unprecedented opportunity to interact with the majestic artist.
Toxic Influence: A Dove Film | Dove Self-Esteem Project
Dove’s “Toxic Influence” marketing campaign uses deepfake to draw attention to the harms of questionable beauty advice spread via social media. The campaign employed mothers of teenage girls who fed them preposterous advice, ostensibly to underscore the adverse effects of influencer culture. The video, created by Ogilvy, also included stunned responses from the misled young women to spread awareness about the unattainable beauty ideals promoted online.
Deepfake Roundtable
Is that even possible? The animation, which features deepfake versions of Robert Downey Jr., George Lucas, Tom Cruise, Ewan McGregor, and Jeff Goldblum, discusses “streaming wars” and how cinema might develop in the future. The show is hosted by Mark Ellis, who moderates the conversation between these iconic filmmakers. The unique opportunity to dramatize the opinions of prominent individuals enables consumers to get a broader understanding of these masters of the film industry and what they might consider the future of the entertainment industry.
Deepfaked Taylor Swift
The video is awesome in manipulating Taylor Swift’s likeness to promote Le Creuset cookware, but it is a scam. Identified from a McAfee advisory, the video has gained traction on social media, which is marked as a deepfake and #AI. The scam merges part of Swift’s body to make her profile relevant to the cookware, exhibiting deceptive use of deepfake technology. McAfee discourages such deceptive clips and urges users to be alert.
Conclusion
To sum up, deepfake technology, in its development and application, has reached far beyond its initial purposes in entertainment. Its use in iconic movies, such as The Shining and Spider-Man: No Way Home or in mainstream media films demonstrates its power and potential. Additionally, it allowed to “resurrect” the legendary celebrities Keanu Reeves and Snoop Dogg to perform on TikTok. Nevertheless, no matter how promising and innovative this technology is, its use in popular mass media and political life is accompanied by numerous ethical considerations and the risk of misuse; therefore, it should be scrutinized and regulated carefully.
The presented deepfake video material demonstrates that, regardless of the inserted message’s level of seriousness or irreverence, the final product makes people question what is true or what is real. Regardless of whether it is a deepfake film material, “deepfake celebrities” performing on the popular TikTok app, used in mainstream news, or adopted for political and social satire create enormous uncertainty around the authenticity of digital expression while focusing heavily on trust.