HomeWhat Are Deepfakes?

What Are Deepfakes?

What are deepfakes and how can they have both positive and negative impact on the world we live in?

Mark
Updated: January 16, 2023

If you’ve seen a video of a celebrity doing something that’s out of the ordinary, chances are, you’re watching a “deepfake” footage. As its name suggests, it’s a forgery, but one that is of higher quality and craftsmanship than other fakeries that it could pass off for the real thing.

Deep fakes are rampant on the Internet right now. Many people use deepfake videos for entertainment purposes. Others also use deep fakes for their political agenda, while others just like to slap their own faces on videos for laughs. Still, there are those that utilize deepfakes for unsavoury intentions – to malign and to slander people, especially those who are popular.

With avatars being a central focal point in the Metaverse, the threat of deepfakes is very real. Since the technology is still in its infancy, now’s the best time to understand this danger and how you can avoid its detrimental effects.

History of the Deepfake

The first practical use of the deepfake surfaced on Reddit in November 2017. At that time,someone unnamed decided to share the algorithm to the platform’s multitude of users. The programming used AI to scan facial biometrics to create convincingly looking humanoids bearing someone else’s face on their bodies.

The first deepfake videos were created by a Reddit user who used AI-based face-swapping technology to create videos that showed celebrities, such as Jennifer Lawrence and Gal Gadot, in pornographic scenes. These videos sparked a lot of controversy and raised concerns about the potential for deep fake technology to be used for malicious purposes, such as spreading false information or harassment.

Unfortunately, the code was meant to be open-source. It meant that several developers could rewrite the code for their purposes, most of which were pornographic in nature. Years later, apps like FaceApp surfaced, which allowed people to alter their own faces or create new individuals using existing facial data.

Notable Positive Uses of the Deepfake

Like all kinds of technology, the deepfake is a double edged sword. It has both its good sides and its bad faces. The technique has been used to great effect in Hollywood films, bringing back actors from the grave to play large roles in certain stories. 

Here are a few notable examples of deepfake technology being maximized in filmmaking:

  • Rogue One: The Star Wars Story

This prequel to the original trilogy shows the story of the Rebel Alliance’s first major conflict with the Galactic Empire. A group led by Cassian Andor (Diego Luna) attempts to steal the plans for the first Death Star, allowing Luke Skywalker’s Red Squadron to discover its weakness. 

You can’t talk Death Star with Darth Vader and Moff Tarkin, played by the late Peter Cushing. Producers molded a digital scan of Cushing’s face onto a proxy actor to shoot the character’s scenes in the film. Another deepfake also appeared at the end of the movie – the face of a young Carrie Fisher imposed on another actress’ face.

  • Furious 7

Furious 7 suffered a period of hiatus beginning November 2013 after one of its main actors died in a tragic car accident. The film’s production was only 50% done, and Paul Walker had just attended a fundraising event in California when he and friend Roger Rodas fatally crashed a 2005 Porsche Carrera.

Although the production team took months to regroup, filming eventually resumed and wrapped up. According to Weta Digital, one of the technologies used to film Walker’s remaining scenes was the deepfake: computer-generated imagery that used the actor’s brothers and archive footage of Walker himself as scanning references. 

  • Captain America: Civil War

The Marvel Cinematic Universe uses all sorts of techniques in visual and special effects to achieve its imagery. Although it’s not a deepfake by definition, the de-aging of Robert Downey Jr.’s Tony Stark in a key scene in Captain America: Civil War makes use of the same technology to achieve the same effect.

The scene showed a younger Tony Stark having an argument with his parents, not knowing that it would be their last interaction. Although they did not replace Downey Jr’s face with someone else, the VFX team did scan the actor’s face and digitally removed all signs of aging to make the final product 20-30 years younger.

Unauthorized Uses of Celebrity Faces Using Deepfake

Photo credit: reAlpha Tech Corp.

With capable and responsible hands, the technology can create various content that plays a positive role in society. However, when used by the wrong parties, they can result in disruptive content that could have serious social and legal repercussions.

For instance, a Russian firm called “Deepcake” recently grafted actor Bruce Willis’ face onto an unnamed actor from the Slavic country for an advertisement. The result was very convincing, but it was unnerving to say the least hearing the actor speak in Russian. It may be amusing to some but the legal aspect is muddy – Willis’ agent has denied giving Deepcake rights to his visage, but a company rep said they claim no ownership as well over any replicas they create.

There are other instances of celebrities’ likenesses being used without their permission. Paperspace Co. recently used the faces of Tom Cruise and Leonardo DiCaprio on their promotional videos. 

Elon Musk, who is currently controversial with his tight-fisted takeover of Twitter, also had his face combined with an unnamed actor for reAlpha Tech Corp.’s marketing clips.

The Dangers of Deepfake Technology

Simply put, the use of deepfake technology to hijack someone else’s face and mould it to another person’s body is synonymous to identity theft. 

In the context of the Metaverse, hackers could steal someone else’s facial biometrics, hand movements and eye-tracking data via the user’s virtual reality headsets and haptic touch gear. This data could be used to program an avatar’s movement and replicate what the original user was doing. 

Hackers can then use the person’s facial biometrics to create a deepfake video. They could also use someone else’s face as well.

These kinds of content can be used, for instance, to slander an individual both on the Metaverse and the real world. Aside from that, identity thieves can also directly threaten the original owner and extort them for money to keep the compromising content private. These individuals rarely honour their commitments, however.

Conclusion

The threat posed by deepfake artists should exert pressure on existing stakeholders of the Metaverse. Before the technology is launched, companies like Mark Zuckerberg’s Meta must find ways to secure their access devices to prevent data theft. Moreover, there needs to be stringent measures in place to verify the identities of everyone on the Metaverse.

Of course, the Metaverse still needs a lot of work before it can become an adaptable technology for most of society. Aside from security issues, the developers all need to figure out how to make each mini-verse compatible and interoperable with one another.

Follow us as we attempt to understand more about the Metaverse as it develops from a nascent technology to become what people call as “Web 3.0”!

Popular Articles

Trending