Deepfakes are one of the more worrying developments when it comes to AI. They are often frighteningly realistic videos of people saying and doing things they have never done before. Generally, they are created by overlaying a face, usually that of a celebrity, over that of another person to create these misleading videos.
They are often meant to be funny, or at least not made with serious intent. Some want to create movies with famous actors (opens in new tab) without ever having to leave the house. But as you can imagine, there is a ton of porn that uses deepfakes (opens in new tab)ranging from fantasies involving video game characters to problematic images of real people.
Aside from the obvious disrespect this can often show towards the images of the people used, one of the most dangerous things about deepfakes is that they’re really good. I’ve seen deepfakes that I never thought weren’t real, and that’s especially dangerous when we have an internet that can spread misinformation like wildfire.
“Deepfake videos are everywhere now. You’ve probably already seen them; Videos of celebrities doing or saying things they never actually did,” explains Ilke Demir, Senior Research Scientist at Intel Labs.
As it turns out, it takes an AI to catch an AI, and luckily Intel has been working on exactly that. The company has been working on efforts as part of its responsible AI work, developing what Intel has dubbed the FakeCatcher. This is a technology specifically designed to detect fake videos like deepfakes and it is said to be able to do it in milliseconds with an accuracy rate of 96%.
Intel explains that many other tools that try to detect deepfakes try to analyze the raw data in the files, while FakeCatcher has been trained using deep learning techniques to recognize what a real human looks like in videos. That means it’s been trained to look for all the little things that make people real. Subtle things like noticing blood flow in pixels of a video that the deepfakes seem to have yet to master. But perhaps they will learn to master our effects on human blood. That’s a scary sentence if I’ve ever written one.
FakeCatcher is said to be working on a web-based platform that will hopefully allow anyone to access the technology. In the back-end, Intel’s 3rd Generation Xeon Scalable processors are used in combination with a suite of proprietary software to perform the necessary calculations.
If you want to learn more about Intel’s FakeCatcher, watch the video at the top of the article. Demir will also be hosting a Twitter Spaces event on November 16th at 11:30am PST where she will further elaborate on the technology being used. Assuming Twitter’s new owner Elon Musk doesn’t fire people (opens in new tab) who ensure that Twitter Spaces work beforehand.