Spotlight

What you need to know about Deepfake?

By codingstreets

Spotlight

Deepfake technology is a type of AI that uses machine learning algorithms to create fake images, videos, audio, and text.

What is a deepfake?

The AI firm Deeptrace found 15,000 deepfake videos online in September 2019, a near doubling over nine months. A staggering 96% were p*rnographic and 99% of those mapped faces from female celebrities on to p*rn stars.

What are they for?

As new techniques allow unskilled people to make deepfakes with a handful of photos, fake videos are likely to spread beyond the celebrity world.

No. Deepfake technology can create convincing but entirely fictional photos from scratch. Audio can be deepfaked too, to create “voice skins” or ”voice clones” of public figures. 

Is it just about videos?

It takes a few steps to make a face-swap video. First, you run thousands of face shots of the two people through an AI algorithm called an encoder. The encoder finds and learns similarities between the two faces, and reduces them to their shared common features, compressing the images in the process.

How are they made?

A second AI algorithm called a decoder is then taught to recover the faces from the compressed images. Because the faces are different, you train one decoder to recover the first person’s face, and another decoder to recover the second person’s face. To perform the face swap, you simply feed encoded images into the “wrong” decoder. 

Everyone from academic and industrial researchers to amateur enthusiasts, visual effects studios and p*rn producers. Governments might be dabbling in the technology, too, as part of their online strategies.

Who is making deepfakes?

It is hard to make a good deepfake on a standard computer. Most are created on high-end desktops with powerful graphics cards or better still with computing power in the cloud. This reduces the processing time from days and weeks to hours. But it takes expertise, too.

What technology do you need?

Poor-quality deepfakes are easier to spot. The lip synching might be bad, or the skin tone patchy. There can be flickering around the edges of transposed faces. And fine details, such as hair, are particularly hard for deepfakes to render well, especially where strands are visible on the fringe.

How do you spot a deepfake?

Badly rendered jewellery and teeth can also be a giveaway, as can strange lighting effects, such as inconsistent illumination and reflections on the iris.

We can expect more deepfakes that harass, intimidate, demean, undermine and destabilise. But will deepfakes spark major international incidents? Here the situation is less clear. A deepfake of a world leader pressing the big red button should not cause armageddon.

Will deepfakes wreak havoc?

Nor will deepfake satellite images of troops massing on a border cause much trouble: most nations have their own reliable security imaging systems.

AI may be the answer. Artificial intelligence already helps to spot fake videos, but many existing detection systems have a serious weakness.

What’s the solution?

They work best for celebrities, because they can train on hours of freely available footage. Tech firms are now working on detection systems that aim to flag up fakes whenever they appear.