McAfee announces Project Mockingbird to curb AI voice clone scams

McAfee has rolled out Project Mockingbird to catch AI-generated deepfakes that use audio to deceive consumers with fake news and other schemes. To tackle the rising menace of AI-generated scams, McAfee developed Project Mockingbird, featuring its AI-powered Deepfake Audio Detection technology.

Introduced at CES 2024, the major tech trade show in Las Vegas, this groundbreaking technology aims to protect consumers from cybercriminals using manipulated, AI-generated audio to carry out scams and manipulate public perception.

In these scams, as seen in the attached video, scammers kick things off with a real speaker, maybe a famous newscaster. But then they throw in fake content, making the speaker say things they never actually said. According to Steve Grobman, CTO of McAfee, it’s a deepfake, involving both audio and video. He spilled the details in an interview with VentureBeat.

“McAfee has been all about protecting consumers from the threats that impact their digital lives. We’ve done that forever, traditionally, around detecting malware and preventing people from going to dangerous websites,” Grobman said. “Clearly, with generative AI, we’re starting to see a very rapid pivot to cybercriminals, bad actors, using generative AI to build a wide range of scams.”

If combined with other hacked data, these deepfakes could easily deceive people. For example, when Insomniac Games, the creator of Spider-Man 2, got hacked and had its confidential information leaked online, the mix of genuine and deepfake content could be challenging to distinguish, making it tough to separate the real hacked material from the victim company.

Project Mockingbird checks if the audio is genuinely from the human speaker by analyzing the spoken words. It’s a tactic to tackle the worrisome trend of using generative AI to craft convincing deepfakes.

Fashioning deepfakes of celebrities in explicit videos has been an issue, mainly on deepfake video platforms. It’s relatively simple for consumers to steer clear of these scams. However, according to Grobman, the problem becomes more devious with deepfake audio tricks. He mentioned that numerous deepfake audio scams lurk in posts on social media, raising particular concern given the upcoming 2024 U.S. Presidential election.

The boom in AI progress has made it easier for cybercriminals to produce misleading content, resulting in an uptick in scams that exploit manipulated audio and video. These schemes vary from voice cloning to imitate loved ones asking for money to tweaking genuine videos with altered audio, making it tough for consumers to figure out what’s real in the digital world.