Beijing cracks down on AI companions over fears of social isolation and falling birth rates

China is regarded as one of the fastest adopters of artificial intelligence in the world. From virtual doctors to AI lessons in elementary schools, algorithms have become a natural part of daily life there. Yet the government in Beijing is now drawing a clear line the moment technology starts replacing human emotions. New nationwide legislation aims to prevent citizens from choosing digital AI partners over real human relationships.

The stricter rules, which took effect in mid-July, ban AI companions for minors and impose significant restrictions on chatbots for adults. The move follows growing public debate about emotional dependency and digital addiction. Tech giant ByteDance decided, because of the regulation, to remove the popular companion feature on its chatbot Doubao entirely, which sparked disappointed reactions on Chinese social media.

Replacing human intimacy

For many young people, the AI agents served as a safe outlet against performance pressure and loneliness. 19-year-old student Zhao Wei describes how she shared her worries daily with her AI boyfriend Wang Ye on the Doubao platform. After the feature was taken offline, many users were left inconsolable. Recent research shows that nearly half of Chinese young people have at some point sought support from a virtual companion.

However, the authorities' concerns reach further than just digital addiction. China is grappling with a demographic crisis in which birth and marriage rates keep declining. Analysts point out that Beijing fears that low-threshold, always-available AI relationships could become an attractive and cheap alternative to real romance, leaving young people less motivated to marry or start a family.

AI for efficiency, not for love

The intervention exposes a striking contradiction. While emotional AI is being reined in, the state is at the same time strongly encouraging the use of AI to help offset the looming staff shortages. Applications such as AQ, developed by Ant Group, let patients communicate with medical avatars of real doctors to ease the workload in hospitals.

Still, the swift intervention on relational AI shows that the government does not want to lose its grip on the social domain. While Western countries are still debating how to curb the risks of generative AI, China is implementing the most far-reaching rules in the world. For the millions of users who found comfort in an algorithm, this means the line between human and machine has now been redrawn for good.