Intel's FakeCatcher technology can instantly detect deepfakes with 96% accuracy

Deepfake technology is one of the most "alarming bells" when it comes to artificial intelligence. The fact is that it is often used to make a video in which a person says certain things that he or she would never say in real life. These are mostly different celebrities or politicians. But there is a way out!

Here's What We Know

Intel is working on FakeCatcher technology, which is able to recognize a video in which DeepFake was used, and it does it in a few milliseconds, and the accuracy is 96%.

Intel explains that many other tools that try to detect DeepFakes try to analyze the raw data in the files, while FakeCatcher is trained to determine what a real person looks like in the video. This means that it is trained to look for all the little things that make people look real.

FakeCatcher is set up to run on a web platform that will hopefully allow anyone to access the technology. On the back end, it uses scalable 3rd generation Intel Xeon processors combined with a bunch of proprietary software to perform the necessary calculations.

Source: PC Gamer