Artificial intelligence (AI) avatars are rapidly gaining prominence, demonstrating their immense commercial potential and challenging the conventional role of human hosts in livestreaming sessions. A recent session on Baidu’s e-commerce platform, Youxuan, generated 55 million yuan in sales.
Luo Yonghao, a prominent livestreamer, confirmed that this was his first experience leveraging virtual human technology for commercial purposes. The AI avatars used were created by Baidu’s generative AI, which analyzed a five-year archive of video content featuring Luo and his co-host, Xiao Mu. This analysis allowed the AI to accurately replicate the hosts’ distinct personalities, mannerisms, and styles, creating highly convincing digital counterparts.
The integration of AI avatars introduces new complexities, including navigating compliance with advertising regulations and adhering to platform rules. Despite these challenges, Luo Yonghao’s team intends to conduct more AI-hosted streams, recognizing the potential for scalability and efficiency. They are also exploring the development of multilingual versions of these AI avatars to expand their reach to global audiences.
This breakthrough is a natural progression within China’s e-commerce landscape, which has pioneered the integration of entertainment and commerce through livestreaming. China’s livestreaming market was already characterized by a massive audience and thousands of human hosts producing an estimated 150,000 hours of content daily. The superior performance demonstrated by AI hosts signals a potential restructuring of the livestreaming labor market, with industry experts noting that AI avatars consistently outperform average livestreamers.
The economic rationale for adopting AI avatars is compelling, as they eliminate the need for costly production teams and dedicated studios, and can operate continuously without breaks. For brands and e-commerce platforms, AI hosts offer the promise of achieving superior results at a lower cost, while mitigating the unpredictability associated with human influencers.
In response to this technological advancement, Chinese authorities are developing regulatory frameworks to address the challenges posed by AI-powered digital humans. Zhejiang Province has introduced guidelines prohibiting the unauthorized use of AI deep synthesis technology and mandating that digital humans be clearly identifiable as non-human entities. Major platforms, such as WeChat, are establishing their own internal policies, with WeChat banning AI-powered digital influencers on its livestreaming platforms.
The Cyberspace Administration of China has proposed comprehensive regulations requiring clear labeling of all AI-generated content to protect consumer interests and prevent misleading practices. These regulatory responses underscore the tension between fostering technological innovation and ensuring trust and transparency within the rapidly evolving livestreaming marketplace.
