McAfee introduces technology to detect deepfakes

Security company McAfee announces Deepfake Audio Detection. The technology is intended for consumers to protect themselves against cybercrime through AI-generated audio. The technology uses various detection models.

Deepfake Audio Detection uses contextual, behavioral, and categorical detection models to determine whether the audio in a video is authentic or likely AI-generated. McAfee’s technology, which is part of its Project Mockingbird, must ensure that consumers are protected against deepfakes. According to McAfee the technology must provide an accuracy of more than 90 percent.

According to McAfee, the tool can be used to recognize scams based on AI-generated audio. Criminals use AI to clone the voices of family members and ask victims to transfer money. The antivirus company also talks about so-called ‘cheapfakes’, in which criminals use AI to manipulate the videos of newsreaders or famous people, for example by making different words come out of someone’s mouth.