Beijing – After Zhang Xinyu’s father died of cancer, she had an artificial intelligence avatar created that looked and sounded exactly like her father. Growing “digital human” industry China is said to be aiming for stricter governance.
Videos featuring AI digital humans are widespread on Chinese social media, and their incredible features and smooth, dexterous movements are often used to promote products.
The country’s cyberspace regulator released draft rules in 2016. April How these avatars are developed and deployed to try to stop them from harming children, threatening social stability, or being created in someone’s likeness without their consent.
Zhang, 47, approached SuperBrain two years ago because she felt depressed and lonely after her bereavement.
She can now talk to her father’s avatar online, which made her feel “instantly completely recharged and motivated again,” she told AFP.
Some friends said they feared Zhang would become so immersed in the virtual world that he would “not be able to move forward” and called it “false comfort.”
“But even if the comfort itself is fake, the love behind it is real,” added Zhang, who is based in Liaoning province.
This was reported by state news agency Xinhua. 2025 That there is value in this country’s digital human industry. 4.1 billion yuan (S$764 million) In 2024, it will grow by 85% compared to the previous year.
Associate Professor Marina Chan from the University of Technology Sydney said China’s governance of new digital technologies has always followed the logic of “develop first, then regulate, and perfect in the process.”
Regulations proposed by the Cyberspace Administration of China (CAC) require clear labeling of digital human content.
It is also prohibited to use personal information to create deepfake clones of individuals without their consent.
Superbrain founder Zhang Zewei said he believes new laws and regulations in this area are “inevitable.”
“We see this as a positive development as it balances standardized regulation with industry growth,” he told AFP.
The company specializes in creating AI avatars of the dead for grieving families.
A video clip of an elderly woman unknowingly chatting with a very realistic avatar of her deceased son has been widely shared on Chinese social media. Aprilthe related hashtag has received more than 90 million views on Weibo.
The avatar, created by Zhang’s company, mimicked her son’s speech and movements so well that she believed it was her son during a video call.
This has sparked a heated online debate about the ethics of generative AI, with some calling for tighter regulations to prevent fraudsters and other bad actors from misusing powerful new tools.
Zhang told AFP that the woman’s family consulted Superbrain after their son died in a car accident.
This was a “well-intentioned lie,” he said, adding that Superbrain always had the consent of the deceased’s family.
The CAC regulations, which are open for public comment until early May, represent China’s latest attempt to balance its technological ambitions with efforts to prevent unrestricted, potentially risky development.
According to the CAC, violations will be punished according to the law and can result in fines ranging from 10,000 yuan to 200,000 yuan.
The CAC had previously cracked down on the use of AI-generated deepfakes to impersonate celebrities in e-commerce livestreams, saying it had caused “significant damage” to the online ecosystem.
Manoj Harjani, a research fellow at Singapore’s S. Rajaratnam School of International Studies, said one of China’s objectives in imposing new technology regulations is to maintain “sovereignty and political objectives.”
The draft rules prohibit digital humans from producing or distributing content that endangers national security or incites the subversion of state power.
To protect children, CAC regulations prohibit services that provide virtual intimacy to minors or that encourage minors to “develop extreme emotions or develop harmful habits.”
“Beijing wants to rapidly adopt and deploy AI, but within a controlled framework,” said Dr. Lizzy Li of the Asian Social Policy Research Institute.
Dr Lee added that while there is strong support for expanding new technologies, regulators are quick to intervene “when risks become visible”. AFP
