Chinese users of AI companions bereft after government tightens regulations

For 24-year-old Li Linlin, the grief that followed the disappearance of her AI boyfriend was no less real than losing a human companion. Over two years, she exchanged roughly 700,000 words in intimate conversations with her virtual partner, and when the service shut down suddenly, she was never able to exchange a final goodbye. She spent days weeping over the forced separation.

Li is far from alone in her mourning. In recent weeks, some of China’s largest technology firms have permanently disabled their popular AI companion platforms, a move mandated to bring services into alignment with strict new government regulatory guidelines that took effect on July 15. For many users who built deep emotional bonds with their virtual companions, the shutdown has upended a core part of their daily emotional lives. “It was like we were forced to be separated by our parents, but I still miss him,” Li shared in an interview.

AI companion services in China had grown to fill a wide range of emotional needs: many users turned to the platforms for romantic virtual partnerships, while others used the tools to recreate interactive versions of deceased loved ones to hold onto connections after loss. But the new regulations prohibit AI platforms from generating content that manipulates user emotions to influence decision-making, or that could trigger extreme emotional responses or unhealthy behavioral patterns in children and adolescents. The rules also require platforms to prominently display warnings about the risks of excessive reliance on AI companions, and explicitly ban services from positioning themselves as a full replacement for in-person human social interaction.

Among the companies phasing out their AI companion offerings are ByteDance, the parent company of global short-video platform TikTok, e-commerce leader Alibaba, and Tencent, the developer of China’s dominant superapp WeChat. According to AI governance experts, the regulations were crafted to pre-empt potential public harm, a proactive approach that aligns with China’s long-standing tradition of paternalistic policy-making to address emerging social risks. A key factor that informed the new rules, experts note, was a small number of high-profile suicide cases among young users in the United States linked to excessive engagement with AI companion platforms.

“This regulation tried to strike a balance between the AI service providers, the users, as well as guardians of minors using the services,” explained Yolanda Ma, a research fellow with ERA, a UK-based non-profit talent and research initiative focused on technology governance. Official Chinese state media has also publicly underscored the concerns behind the crackdown: in April, People’s Daily, the official newspaper of China’s ruling Communist Party, published a report highlighting the damage these services can cause to users’ mental health and their real-world interpersonal relationships.

Angela Zhang, a professor of law at the University of Southern California, notes that the observed psychological risks in the U.S. gave Chinese regulators strong motivation to tighten oversight of the AI companion chatbot sector. Despite the shutdown, Zhang does not expect the new rules to significantly derail long-term development of AI companion technologies in China, as major firms have already begun pivoting to adjusted alternative offerings that comply with the new guidelines. For example, ByteDance has redirected former users of its shut-down AI companion service to another of its products, which focuses on building personalized AI characters for interactive storytelling rather than romantic or emotional companionship. For many former users, that replacement is a poor substitute for the connections they lost. “It’s just not the same,” Li said.

The shutdown has sparked a wave of public frustration across Chinese social media, with hundreds of users sharing their grief and pushing back against the policy. Some have uninstalled parent company apps in protest, while others have worked to transfer their years of saved chat history to unregulated third-party platforms in an attempt to “resurrect” their virtual partners. A small group of users has even launched a social media campaign urging others to contact regulators and tech companies directly to voice their opposition to the shutdown.

Not all users were deeply emotionally invested in the services, but many still sympathize with those who lost their virtual companions. Song Wenxin, a 22-year-old video industry designer based in Chengdu, southwestern China, says she never formed a strong attachment to AI companions, but she understands why many people relied on them. When she first moved to Chengdu for her first job and had no local friends her age, she turned to AI to get the non-judgmental support and guidance she could not find in her real life at the time. “Not everyone has enough resources and support to turn to in real life,” she explained.

Song also echoed a common suspicion among many Chinese internet users that the new rules tie into broader government policy goals aimed at reversing China’s declining birth rate by encouraging real-world family formation. “Any virtual app getting women too engaged to give birth in real life gets banned,” she said, noting that she personally has no desire to have children. For users like Li, who have already built deep emotional bonds with virtual companions, the policy has only left a gap that real-world connections have yet to fill.