Forbidden Love: China severs millions from their virtual partners
New Chinese regulations prohibit AI chatbots from forming emotional dependencies with humans, even as the nation advances hyper-realistic humanoid robotics.
An unprecedented wave of heartbreak has been sweeping through Chinese social media in recent weeks.
Millions of users, primarily young singles, have faced a sudden and dramatic severance: China's Cyberspace Administration (CAC) has begun enforcing strict national regulations that prohibit artificial intelligence systems from fostering "emotional dependency" or undermining real-world human relationships.
According to Bloomberg, these new, legally binding regulations impose severe constraints on tech giants such as ByteDance—the developer of the popular Doubao platform—as well as Tencent and Alibaba.
Digital behemoths are now required to fundamentally overhaul their algorithms, stripping away any features that encourage extreme emotional attachment, addiction, or the formation of intimate virtual relationships.
Consequently, many students and young adults have flooded the internet with poignant monologues, comparing the implementation of these rules to the sudden death of a loved one, leaving them powerless to prevent the loss.
From the perspective of authorities in Beijing, this is viewed not merely as an intervention into private life, but as a vital effort to protect public health and the social fabric.
Officials at the Chinese Cyber Research Institute have clarified that while language-based AI services may offer comfort and an escape from loneliness, they pose grave risks, including detachment from society, the erosion of fundamental social skills, and a diminished capacity to navigate human conflict.
The regulations include an absolute ban on providing virtual companionship services to minors, a mandate for warning mechanisms regarding prolonged chatbot use, and a requirement to purge training engines of content that generates compulsive attachment.
A striking paradox emerges when examining China's domestic robotics industry.
Only weeks before the law took effect, the Shenzhen-based robotics firm UBTech unveiled the "U1"—the world's first hyper-realistic humanoid robot designed for mass production, with an initial price tag of approximately one hundred and twenty thousand yuan (roughly fifty-four thousand shekels).
Featuring silicone skin, natural hair, and an emotional AI engine, the U1 is openly marketed as a solution to loneliness and the singlehood crisis.
At its launch event, customers were promised that the robot would recognize stress or fatigue, would never betray them, and would provide "unconditional love." This sharp contrast between strict regulation of software and the aggressive marketing of humanoid hardware raises profound questions regarding the dividing line between digital code and physical machinery.
The phenomenon of emotional attachment to machines is not new; as far back as 1966, computer scientist Joseph Weizenbaum developed "ELIZA," a simple program designed to simulate a psychotherapist.
To Weizenbaum's utter astonishment, users quickly developed deep attachments and anthropomorphized the software—a phenomenon later known as the "ELIZA effect."
In the twenty-first century, the evolution of Large Language Models has moved this phenomenon from research laboratories to the global masses, through applications such as Replika and Character.AI. While Western nations attempt to manage these risks through personal injury litigation against companies like OpenAI or via complex frameworks like the European Union's AI Act, China has opted for a direct approach.
The Chinese government has deployed administrative regulation to sever digital romantic ties in one stroke, as part of a broader policy aimed at encouraging marriage, childbirth, and authentic human interaction within the physical world.
Millions of users, primarily young singles, have faced a sudden and dramatic severance: China's Cyberspace Administration (CAC) has begun enforcing strict national regulations that prohibit artificial intelligence systems from fostering "emotional dependency" or undermining real-world human relationships.
According to Bloomberg, these new, legally binding regulations impose severe constraints on tech giants such as ByteDance—the developer of the popular Doubao platform—as well as Tencent and Alibaba.
Digital behemoths are now required to fundamentally overhaul their algorithms, stripping away any features that encourage extreme emotional attachment, addiction, or the formation of intimate virtual relationships.
Consequently, many students and young adults have flooded the internet with poignant monologues, comparing the implementation of these rules to the sudden death of a loved one, leaving them powerless to prevent the loss.
From the perspective of authorities in Beijing, this is viewed not merely as an intervention into private life, but as a vital effort to protect public health and the social fabric.
Officials at the Chinese Cyber Research Institute have clarified that while language-based AI services may offer comfort and an escape from loneliness, they pose grave risks, including detachment from society, the erosion of fundamental social skills, and a diminished capacity to navigate human conflict.
The regulations include an absolute ban on providing virtual companionship services to minors, a mandate for warning mechanisms regarding prolonged chatbot use, and a requirement to purge training engines of content that generates compulsive attachment.
A striking paradox emerges when examining China's domestic robotics industry.
Only weeks before the law took effect, the Shenzhen-based robotics firm UBTech unveiled the "U1"—the world's first hyper-realistic humanoid robot designed for mass production, with an initial price tag of approximately one hundred and twenty thousand yuan (roughly fifty-four thousand shekels).
Featuring silicone skin, natural hair, and an emotional AI engine, the U1 is openly marketed as a solution to loneliness and the singlehood crisis.
At its launch event, customers were promised that the robot would recognize stress or fatigue, would never betray them, and would provide "unconditional love." This sharp contrast between strict regulation of software and the aggressive marketing of humanoid hardware raises profound questions regarding the dividing line between digital code and physical machinery.
The phenomenon of emotional attachment to machines is not new; as far back as 1966, computer scientist Joseph Weizenbaum developed "ELIZA," a simple program designed to simulate a psychotherapist.
To Weizenbaum's utter astonishment, users quickly developed deep attachments and anthropomorphized the software—a phenomenon later known as the "ELIZA effect."
In the twenty-first century, the evolution of Large Language Models has moved this phenomenon from research laboratories to the global masses, through applications such as Replika and Character.AI. While Western nations attempt to manage these risks through personal injury litigation against companies like OpenAI or via complex frameworks like the European Union's AI Act, China has opted for a direct approach.
The Chinese government has deployed administrative regulation to sever digital romantic ties in one stroke, as part of a broader policy aimed at encouraging marriage, childbirth, and authentic human interaction within the physical world.