© 2023 CyberNews - Latest tech news,
product reviews, and analyses.

If you purchase via links on our site, we may receive affiliate commissions.

China’s new deepfake regulations aim to protect the regime, but at what cost?


In what is, in essence, official advice on how to do deepfakes the right way, China’s cyberspace regulator announced that new rules for content providers that alter facial and voice data will take effect from January 10.

Cyberspace Administration of China (CAC) looks to more tightly scrutinize the so-called “deepfake” technology and services and its regulations, issued late on Sunday, to protect people from being impersonated.

The latter images are virtually indistinguishable from the original and easily used for manipulation or misinformation. Deepfakes are controversial outside China for their potential to mislead audiences, and Beijing clearly also has worries about the technique.

The CAC regulations prohibit the creation of deepfakes without the subject’s permission and the depiction of anything that could be considered contrary to the national interest – according to observers, anything not socialist enough would fall under that description.

The regulator said the move was aimed at curbing risks that might arise from activities provided by “deep synthesis service providers” that use deep learning or virtual reality to alter digital content.

However, the regulations also suggest that China wants to promote the industry's healthy development and actually expects deepfakes to be widely used – under Beijing's supervision.

Deepfakes must “promote the healthy development of internet information services and maintain a good ecology of cyberspace,” the statement on the CAC website says.

Technical problems

China’s government has been increasingly relying on tech companies to enforce new Internet regulations to facilitate the Chinese Communist Party’s vision of a stable, prosperous society, Emmie Hine and Luciano Floridi recently wrote in a comment on Nature Machine Intelligence.

To these researchers, the issued regulations pose some technical problems. For example, “synthetic content” must be labeled, yet it’s not clear how to ensure that labels are created and preserved – even whole-frame watermarks can be removed by re-encoding or using another AI system.

The new rules also include a requirement for the registration of users – including their real names. Real-name verification laws will certainly enable authorities to track the sources and spreaders of unlabeled or misleading deepfakes.

“But this is challenging because, once a piece of content is created, it can be decoupled from the service on which it is created and spread independently of it. Videos can be re-uploaded, audio re-recorded, images screenshotted, removing them from the originator’s control,” Hine and Floridi wrote.

“And, as the adage goes, the Internet is forever: once some information has spread on the Internet, it is extremely difficult to erase it completely.”

Social (in)stability

The need to control the internet stems from the Chinese government’s wish to develop “social stability” under its own rules. To Beijing, an open web ecosystem is a destabilizing factor – as are deep synthesis services or deepfakes.

“For a government that relies on being able to control what information is available to its citizens, synthetic content represents an existential threat.”

Emmie Hine and Luciano Floridi.

“A single fake post can be censored or officially discredited. However, a hypothetical fake conspiracy theory supported by deepfaked photographs, videos and documents may not be countered so easily. Even when content is removed from the Internet, it lingers in people’s minds,” Hine and Floridi said.

“For a government that relies on being able to control what information is available to its citizens, synthetic content represents an existential threat.”

According to researchers, deepfakes can be spread more easily than reliable news, which can be more easily blocked. That’s supposedly why China is moving to regulate deepfake technology more quickly and thoroughly than the United States or the European Union.

Only a handful of American states have issued regulations for deep synthesis technology, and those focus solely on deepfake pornography or deepfakes intended to influence elections.

The EU has moved to require deepfake labeling from platforms through amendments to the Digital Services Act, which would take effect in 2023.


More from Cybernews:

Weekly recap: from lawsuit against TikTok to viral Chat GPT

Cybersecurity boss pleads guilty to defrauding own company

Android app with over 5m downloads leaked user browsing history

Google told to remove search results about users

Dyson releasing cyberpunk anti-pollution headphones for $949 a pair

Subscribe to our newsletter



Leave a Reply

Your email address will not be published. Required fields are marked