What is a Deepfake?

In an era of misinformation legit deepfakes are coming, and they might already be here and we wouldn’t know the difference. That is, “perfectly realistic” deepfakes aren’t far off and as our brains get used to them, that “purely false information” will simply become our new reality. The internet is reaching a stage of virtual reality that doesn’t correlate ot translate with “reality” all too well.

Deepfakes refer to manipulated videos, or other digital representations produced by sophisticated artificial intelligence, that yield fabricated images and sounds that appear to be real.

The word deepfake combines the terms “deep learning” and “fake,” and is a form of artificial intelligence.

From Instagram profiles to TikTok skits, there’s something already unreal and meme-like about how the internet is evolving.

The written word is being lost for the micro video.

But the Deepfake is the ultimate riddle of the video world, where we can’t know what is real or what is fake.

A few years ago, deepfake technology would be limited to use by nation states with the resources and advanced technology needed to develop such a powerful tool. Now, because deepfake toolkits are freely available and easy to learn, anyone with internet access, time and motive can churn out deepfake videos in real time and flood social media channels with fake content.

Any company with a background in AI and the biometric databases of citizens can create deepfakes of us that would be reliable and believable by people we know. Camera apps have become increasingly sophisticated. Deep learning is also contributing to a whole new niche of content that could be misleading, prone to the spread of misinformation and what potentially goes viral in the meme-like quality of our skim attention.

Most deepfake technology is based on generative adversarial networks (GANs). Now, a video can be fabricated from a single photo. In the future, where all it takes is your Facebook profile image and an audio soundbite of you from an Instagram story, everybody becomes a target.

The reality of information in a democracy becomes a caricature of what the media used to be.

What is a Deepfake Video?

What happens in a world where computers can generate convincing representations of events that never happened? Our identities can be stolen, our actions can be framed, our reputations can be warped. It’s like a new cybersecurity and misinformation threat.

GANs enable algorithms to move beyond classifying data into generating or creating images. This occurs when two GANs try to fool each other into thinking an image is “real.” Deep learning too has its own peculiar risks as it evolves and our entire online identities intersect with the new technology.

“Deepfakes” are videos that use AI technology to paste a celebrity face onto a different body.
— Sarah Rense, Esquire, 12 Feb. 2018

Even what a Deepfake is since 2018 has changed. Deepfakes are not just for celebrities anymore in 2020.

At the Last Futurist when we talk about the dangers of AI and its lack of regulation, deepfakes are yet another emerging field. This is because Artificial intelligence effectively can learn what a source face looks like at different angles in order to transpose the face onto a target, usually an actor, as if it were a mask.

These huge advances came through the application of generative adversarial networks (GANS) to pit two AI algorithms against each other, one creating the fakes and the other grading its efforts, teaching the synthesis engine to make better forgeries.

We can no longer trust our senses in a world of AI. Think about it, Deepfakes bring AI that can deceive us, as if the internet were not already a dishonest and unreliable place.

The Reality of Non-Reality

Deepfakes are so powerful because they subvert a basic human understanding of reality: if you see it and hear it, it must be real. Deepfakes untether truth from reality. This is problematic for people because we are used to relying on our senses to better understand reality and trust sources of information. While GenZ and Millennials who grew up with the internet are skilled at ignoring Ads and banner ads, deepfakes are a whole different story.

Soon you will need AI to tell if something is real or not, and even that could be prone to error.

Last October, Governor Gavin Newsom signed California’s AB 730, known as the “Anti-Deepfake Bill,” into law with the intent to quell the spread of malicious deepfakes before the 2020 election. However if the internet and BigTech are not regulated, how can deepfakes be properly regulated in the 2020s?

Are Deepfakes Dangerous?

The vast amount of fake news that already exists foreshadows what could happen as deepfakes become easier to create at scale. It’s hard to impossible to regulate the proliferation of video content, both real and fake. This as platforms like TikTok, Facebook and YouTube are teaching us.

Even in January, 2020 and in recent weeks, Facebook, Twitter and TikTok have all found themselves scrambling to tackle this new form of disinformation: deepfake videos. We already live in a world of memes, glimpse impressions, first impressions and superficial exposure and consumption of online experiences in the scale of seconds. Deep fakes could be a game changer.

We increasingly find things that are not true, amusing. From TikTok to viral memes, we already live in a world of distortions, illusions and partially immersive virtual reality. From AR to deepfakes, this silly world is becoming more distinct from facts, news and an internet without pressure from advertising, app engagement and varied attempts to get eyeballs and retain our mobile attention.

Deepfake manipulation is also an emerging cybersecurity threat. This is because the human layer is always the most vulnerable. Spearphishing targets high-level employees, typically to trick them into completing a manual task such as paying a fake invoice, sending physical documents or manually resetting a user’s credentials for the cybercriminal.

Deepfakes include videos manipulated using artificial intelligence to make it look like someone is saying something they actually haven’t. However it’s a stream of the internet that’s only going to become more real, and that’s not a good thing.

In simplistic terms, deepfakes are falsified videos made by means of deep learning. If Deep learning is “a subset of AI,” and refers to arrangements of algorithms that can learn and make intelligent decisions on their own, clearly the evolution of the “deepfake” bears some watching.

If you want to talk about the impact of AI on society, we’ll have to get our heads around these deep-learning systems of persuasive counterfeit realities, meme-layers of illusions and fraud. As if the advertising based internet were’t already somewhat based on deception and clickbait.

AI doesn’t just augment our humanity, it augments our capacity for exploiting each other, creating not always a more orderly society. Deepfakes need to be regulated, just as many aspects of technology and AI that don’t currently have proper rules, oversight, legal frameworks and ethical bodies.