The unauthorized use of AI to create and spread deepfakes impersonating celebrities, including movie icon Stephen Chow, has ignited a legal and ethical debate in China, prompting calls for stricter platform governance amid broader concerns over digital identity rights, China Media Group reported.

On Monday, Chow's agent Chen Zhenyu questioned the legality of these videos in a Weibo post, suggesting that creators likely profited from them and criticizing the platform for allegedly failing to regulate their distribution.

Meanwhile, as Spring Festival approaches, unauthorized AI-generated greeting videos featuring celebrities have also stirred controversy online. Some clips depict comedians like Feng Gong and Cai Ming, along with Hong Kong star Andy Lau, appearing in users' homes where they are making dumplings and extending holiday wishes.

Zhang Linghan, director of the Institute of AI Law and Governance at the China University of Political Science and Law, said facial and vocal data are sensitive personal information with biometric functions. Unauthorized deepfakes violate the subject's rights.

Under China's Civil Code and regulations on deep synthesis, faces and voices are classified as sensitive personal data. Unauthorized creation or dissemination infringes on portrait and voice rights, with liabilities applying even for non-commercial use. Highly realistic synthetic content blurs the line between truth and fabrication, risking misuse for rumors or fraud and endangering the online ecosystem.