DIGITAL LIFE

Can You Spot A Face Swap? Take The Deepfake Test

There is no doubting that cyberattacks are growing ever more sophisticated and dangerous, with warnings from the FBI to mitigate the phishing threat with two-factor authentication, and infostealer logs holding millions of stolen passwords. One area where that sophistication is there in plain view for everyone to see is fraud, specifically the use of face-swapping attacks, which have surged by 300% over the last 12 months, according to new threat intelligence. But can you spot a deepfake? There’s one way to find out, take the deepfake test.

Deepfake Face Swap Attacks Are Surging...The annual iProov Threat Intelligence Report, published Feb. 27, has confirmed what most infosecurity professionals already suspected: identity attacks fueled by the increasing sophistication of AI-powered deepfake technology are surging. By analyzing global cybercrime activity based on the live observations provided by the iProov Security Operations Center, threat intelligence experts were able to determine that attacks using synthetic, AI-powered, imagery known as face swaps surged by 300% over the course of the last year. This, coupled with the 2,665% increase in the use of native virtual camera attacks, where a photo is used to replace a live selfie used for identity verification purposes, shows the direction of travel as far as fraud is concerned. Throw the use of image-to-video conversion, which iProov analysts warned “has emerged as a new synthetic identity attack vector with a simple, two-step process that could impact the effectiveness of many liveness detection solutions on the market,” and the seriousness of the deepfake threat is apparent to all.

"The commoditization and commercialization of deepfakes, for instance, pose a significant threat to organizations and individuals,” Dr. Andrew Newell, chief scientific officer at iProov., said; “What was once the domain of high-skilled actors has been transformed by an accessible marketplace of tools and services that low-skilled actors can now use with minimal technical expertise for maximum results.”

Can You Spot A Deepfake?...According to an iProov study of some 2,000 consumers in the U.S. and U.K., only 0.1% could accurately spot a deepfake generated by AI. Videos are the trickiest to spot, with people being 36% less likely to spot a deepfake when compared to a still image. Bizarrely, though, 60% of those asked were convinced in their confidence to be able to detect these face swap attack methods. “This study shows that organizations can no longer rely on human judgment to spot deepfakes and must look to alternative means of authenticating the users of their systems and services,” Professor Edgar Whitley, a digital identity expert at the London School of Economics and Political Science, said. Which begs the question, can you spot a deepfake?

Take the test here--https://quiz.iproov.com/

“Criminals are exploiting consumers’ inability to distinguish real from fake imagery, putting our personal information and financial security at risk,” Andrew Bud, founder and CEO of iProov, said; “It’s down to technology companies to protect their customers by implementing robust security measures.” Let’s face it, attackers are not going to backtrack on the deepfake threat while it remains so successful, leaving a combination of user awareness and provider protections as the only defense.

Davey Winder