Going Deep Into DeepFakes – Part 1 – What the Heck is Going On

Going Deep Into DeepFakes – Part 1 – What the Heck is Going On

If you’ve been paying any attention at all to what’s going on, you must have heard about DeepFakes. In case you happened to have not, DeepFake is a technology that uses Deep Learning (a promising… and delivering!… breakthrough technique in Artificial Intelligence) to take an existing video (e.g., a scene from a movie, an interview, an Oscar awards thank you speech, a political debate or your personal video) and convincingly overlay the image of a different person into it so that it looks like its the new person that was filmed. And I want to iterate convincingly. So convincingly, in fact, that politicians are scared of the impact of DeepFakes. They are scared that someone will deliver a subversive message using their appearance and all watchers will be none-the-wiser. In fact, you can see an example for yourself:

Just like that, with a snap of the fingers, even several years ago, people could already make a video of the former president saying anything they wanted. If you would like to learn how to do something like this, you can enroll in Machine Learning for Red Team Hackers.

Now, it’s been several years since this video. What else has been happening?

In this little post, I’m going to catch you up on visual DeepFakes. It is a part of a 4-part series on DeepFakes, with the next 3 posts being

  • Going Deep Into DeepFakes – Part 2 – Don’t Believe Everything You Hear
  • Going Deep Into DeepFakes – Part 3 – AI-generated Reviews, Weaponizing Twitter and Artifically-Generated Universes
  • Going Deep Into DeepFakes – Part 4 – How Humanity Can Persevere Against DeepFakes

Before I go there, I want to share with you results of a survey I recently conducted via Linkedin. It’s not a massive survey and it’s mostly tech savvy folk, so take it with a grain of salt, but here the results:

In other words, most people that are tech savvy are worried about DeepFakes and have at least heard of it or seen a demo of the technology.

Now let’s see what’s been happening in the past few years in the universe of “traditional”, visual DeepFakes.

Politics

When it comes to real life political applications of the technology, the widespread fears seem to be getting credence. DeepFake technology does persuade the general population.

A recent political application took place in April 2020, where the Belgian branch of Extinction Rebellion published a deepfake video of Belgian Prime Minister Sophie Wilmès on Facebook. The video promoted a possible link between deforestation and COVID-19. It had more than 100,000 views within 24 hours and received many comments. On the Facebook page where the video appeared, many users interpreted the deepfake video as genuine.

On the other hand, positive political applications exist as well. During the 2020 Delhi Legislative Assembly election campaign, the Delhi Bharatiya Janata Party used the technology to distribute a version of an English-language campaign advertisement by its leader, Manoj Tiwari, translated into Haryanvi to target Haryana voters. A voiceover was provided by an actor, and AI trained using video of Tiwari speeches was used to lip-sync the video to the new voiceover. In this case, DeepFake allowed the party to convincingly approach the target audience even if the candidate didn’t speak the language of the voter. Sweet.

Porn

Well, if you know people, you’d know that the technology is being extensively used for… you guessed it, porn. I’m not going to delve deeply into this, except to say that the technology is able to uncloth people, swap-in celebrity faces into existing videos and be used to humiliate targets by putting their faces on fake videos. One more thing to note is that many websites, such as Pornhub, Reddit and Twitter are banning such videos. But to ban a DeepFake, you must first identify it as a DeepFake. And that’s hard and getting harder.

Entertainment

Moving on after washing my hands from this topic, much of DeepFake technology is used for entertainment purposes. There are apps like FakeApp, Faceswap, DeepFaceLab, Zao that allow ordinary persons with no knowledge of the inner working of the technology to create DeepFakes.

You too can download such an app and create a DeepFake for your face. I show how you can do that in Machine Learning for Red Team Hackers.

Art

DeepFakes can now bring old art back to life. If you look around, you can find singing videos of Einstein, conversations with the Mona Lisa and modern celebrities performing in unexpected music videos.

Cinema

DeepFake technology is also being used to create more convincing cinematic experiences. For example, in “Solo: A Star Wars Story”, Ford’s young face was inserted into his character. A similar technique was used for Princess Leia in “Rogue One”.

Fake Personas

On a more subversive level, DeepFakes can be used to create online “sockpuppets”, which are fake personas used by hidden puppet-masters with an agenda. A person named Oliver Taylor, whose identity was described as a university student in the United Kingdom, submitted opinion pieces in several newspapers and was active in online media attacking British legal academic Mazen Masri and his wife Ryvka Barnard, as “terrorist symphathizers.” Reuters could find only scant records for Oliver Taylor and “his” university had no records for him. Many experts agreed that “his” photo is a deepfake. Several newspapers have not retracted “his” articles or removed them from their websites. It is feared that such techniques are a new battleground in disinformation.

What Now

Cool, so now you’ve gotten a glimpse of recent DeepFake developments. “What should I make of this?”, you might say. Several things. First, don’t believe everything you see. Just because you saw some politician saying something in a video, doesn’t mean it’s real. Tell this also to your close ones, since they probably know less than you do about DeepFakes and technology.

Secondly, it’s obvious that DeepFakes are going to be critical in the future economy. Movies will employ DeepFakes. Chatbots will employ DeepFakes. DeepFake will be used in post-production, like make-up artists touch up actors nowdays. Entertainment will leverage DeepFakes. And politics will require serious Cybersecurity knowledge of DeepFakes. So if you want to be  ahead of the curve, on the frontier of the future economy, I suggest you learn about how to actually make DeepFakes. By enrolling in the highly-rated Machine Learning for Red Team Hackers course, you can learn just that.

In the next post, I’m going to cover audio DeepFake, aka voice transfer technology. Subscribe at ML4CS.com to make sure you don’t miss it.