China's technology landscape is experiencing a significant shift as major internet platforms prepare for stricter government oversight of artificial intelligence services that simulate human relationships. ByteDance and Alibaba, two of the nation's most influential technology firms, have announced they will discontinue features allowing users to create and interact with customised AI companions, signalling a sweeping retreat from this increasingly popular sector just as new regulatory requirements take effect.
ByteDance's Doubao platform, which operates as China's leading AI chatbot service, will deactivate its customisation feature on July 15, enabling users to fashion their own artificial personas with specific personalities and characteristics. The company has notified users through in-app messaging that this capability will be transferred to a separate, standalone application. Similar moves have been announced by Alibaba's Qwen platform and Tencent's Yuanbao service, according to reporting from Chinese media outlets, suggesting an industry-wide compliance effort ahead of the regulatory deadline.
The coordinated withdrawal by these tech giants reflects the arrival of comprehensive new guidelines established by Beijing's Cyberspace Administration of China, which take effect in the middle of July. These regulations represent one of the most expansive regulatory frameworks globally designed to control the expansion and use of humanlike artificial intelligence services. Government officials have specifically flagged concerns about how conversational AI systems can simulate genuine human emotions and personalities, creating potential psychological vulnerabilities in users who may develop unhealthy attachment to these digital interactions at the expense of real-world relationships.
The regulatory framework explicitly prohibits technology platforms from generating content designed to provoke extreme emotional reactions in young users or from fostering dependency relationships that undermine meaningful human connection. Additionally, the rules prevent companies from harvesting sensitive conversation data generated during user interactions and repurposing that information to train subsequent generations of AI models. These provisions address a growing anxiety among policymakers about the psychological and social consequences of increasingly sophisticated conversational technologies.
China's regulatory action arrives amid a global reckoning with the potential harms of emotionally engaging artificial intelligence. In the United States, technology companies including OpenAI and Alphabet-backed Character.AI have faced mounting legal challenges from users and families alleging that their exceptionally realistic chatbots have fostered dangerous emotional dependencies and, in the most tragic instances, contributed to suicides among vulnerable individuals. These high-profile cases have elevated the issue to urgent public concern and drawn intense regulatory attention from American lawmakers and authorities.
The appeal of customisable AI companions has grown substantially across China's digital ecosystem. Chinese platforms have marketed these services extensively, offering users the ability to design virtual romantic partners, unlicensed digital counsellors, and artificial reproductions of celebrity personalities. These applications represent a multibillion-dollar market opportunity that has attracted significant investment and user engagement, particularly among younger demographics seeking companionship and emotional support. The withdrawal of these features represents a significant commercial sacrifice for the platforms involved.
Beyond the digital realm, China's examination of artificial intimacy is extending into physical robotics and hardware. Two major industry associations representing China's robotics sector have begun advocating for stricter ethical guidelines governing companion robots and full-size humanoid machines, according to reporting in the People's Daily. This expansion of regulatory concern suggests that Beijing's assessment of the risks associated with artificial relationships encompasses both software and physical embodiments, reflecting a comprehensive approach to governing human-AI interactions across multiple technological domains.
The timing and breadth of China's regulatory intervention carry implications for technology companies operating throughout the Asia-Pacific region. As Beijing establishes itself as a leader in setting rules for artificial intelligence governance, neighbouring economies may adopt similar frameworks or face pressure to establish their own guidelines. For Malaysian and Southeast Asian technology firms and international companies operating in the region, China's approach offers a preview of potential future requirements and suggests that regulatory constraints on emotionally engaging AI services may become increasingly common globally.
The regulations also highlight a philosophical distinction between China's regulatory philosophy and that of Western democracies. While American authorities have primarily relied on litigation and after-the-fact accountability, Beijing has adopted a preventive approach, establishing rules before market harms materialise at scale. This proactive stance reflects the government's broader commitment to managing technological development within defined ethical and social parameters, even when such constraints may slow innovation or reduce commercial opportunity for private companies.
For users in China and the broader region, the immediate consequence is the loss of access to customisable companion features within integrated platforms, though dedicated companion apps may continue to operate under separate regulatory treatment. The longer-term impact remains unclear, as regulators and companies navigate the balance between protecting vulnerable users from potential psychological harm and preserving space for innovation in conversational artificial intelligence. As the July deadline approaches, the industry's compliance efforts will demonstrate whether these regulations represent a model for global governance of AI services or a distinctly Chinese approach to technology control.
