Tik Tok has denied developing deepfake technology on their app, and to prove it, they’ve deleted the code that existed within their app.

 

 

TikTok, the world’s fastest-growing social media platform uber-popular with children, has had life-like deepfake technology developed for it by its parent company, ByteDance. 

The feature allows users of TikTok and its Chinese sister app, Douyin, to insert their face onto any number of videos, requiring only a multi-angle biometric scan of one’s face.

Tech companies trying to consumerise weak version of deepfakes is not new, with plenty of face-swap and face-rendering apps on the market already.

However, none hold a candle to ByteDance’s convincingly life-like deepfake technology that is set to be released on its flagship applications that have already totalled over 1.5 billion downloads.

Mobil research startup Watchful.ai discovered code for the deepfakes feature in the framework of the latest update of the apps. After tipping off TechCrunch, TechCrunch reached out to TikTok and Douyin for comment on the feature.

TikTok declined to answer any questions posed but a spokesperson has denied the existence of the feature and said TikTok has no intention of introducing it. They later told TechCrunch that “…the inactive code fragments are being removed to eliminate any confusion,” implicitly confirming deepfake code was present in TikTok’s application. 

From what Watchful could gather, ByteDance has set some responsible safeguards on the technology. At the very least, the feature will be restricted to those of age and the biometric scan also serves as an identity check, requiring the user to make a variety of face actions, to ensure that a non-consenting individual is not being used.

 

Users also cannot upload their own videos and have to make do with the ones provided within the app, negating the possibility of non-consensual pornography and other misinforming videos.

 

Users also cannot upload their own videos and have to make do with the ones provided within the app, negating the possibility of non-consensual pornography and other misinforming videos.

It’s possible that the technology never reaches the market, but that doesn’t change the fact that it has been developed and is functional. It illustrates ByteDance’s willingness to embrace the controversial technology and opens up many possibilities for not only ByteDance but other companies and groups that could potentially develop their own version of it.

It is also possible that a relatively tame deepfake release on TikTok and Douyin sets a less extreme and more convivial scene for the technology where it is used fun and entertainment instead of the spread of misinformation. It may also tune people’s perception of internet information, heightening their awareness that they cannot believe everything they see online. Regardless, it begs the question: what can ByteDance do with such sensitive biometric data?

Watchful.ai co-founder and CEO Itay Kahana is not so optimistic.

“These deepfake apps might seem like fun on the surface, but they should not be allowed to become trojan horses, compromising IP rights and personal data, especially personal data from minors who are overwhelmingly the heaviest users of TikTok to date,” they said. 

 

 

 

 

Share via