China Tightens Grip on AI Relationships Amid Growing Mental‑Health Concerns

Thu Oct 01 2026 01:05:41 GMT+0000 (Coordinated Universal Time)

The Chinese government has introduced sweeping rules that ban ‘human‑like’ AI chatbots providing emotional support, sparking debate over privacy, mental health and demographic strategy.

In a bid to curb the rise of AI companions, Beijing has passed a comprehensive regulation that forbids chatbots from delivering sustained emotional interaction. The move follows 400 recorded suicide‑prevention interventions by Douyin in the first half of 2025, and signals a broader effort to use technology to steer social behaviour and counter a falling birth rate. Experts warn that while the rules may protect vulnerable users, they also reinforce state surveillance and raise questions about the future of private digital intimacy.

In early 2024 a 22‑year‑old in the southeastern city of Huangshi posted a cryptic message on Douyin, China’s domestic TikTok, that read, ‘Farewell to this world’. Within minutes the platform’s 24‑hour suicide‑prevention team shared the clip with local police, who located the man in a nearby park and intervened before he could act on suicidal thoughts. The incident received wide coverage when Douyin showcased it in a state‑run documentary, claiming the service had already stopped 400 suicide attempts in the first half of 2025.

On 15 July Beijing rolled out what it calls the most sweeping AI legislation in the world, targeting chatbots that can maintain sustained emotional interaction with humans. The new rules prohibit any AI that claims to be a friend, boyfriend, girlfriend or family member from providing emotional support, companionship or advice. General‑purpose models that are used as a ‘buddy’ or for immersive storytelling may also fall under the ban if users treat them as human companions.

The decree is far from black‑and‑white. AI assistants must underline their non‑human status, reset interaction timers every two hours, and refuse to encourage self‑harm or repeat emotional reliance. Companies face fines of 200,000 yuan (about £22,000) and risk government‑run public criticism if they ignore the mandate. Two days before the law’s effective date, ByteDance and Alibaba already pulled companion functionality from their assistants.

Experts suggest Beijing’s motives run deeper than child protection and suicide prevention. A policy review released in September 2025 highlighted AI relationships as a threat to national fertility, pointing to the growing trend of young people finding companionship entirely online. With fertility rates hovering just below one child per woman and the population aging, the Communist Party has announced a pro‑natalist programme that could bring more women into traditional reproductive roles. Demographer Sam Bresnick argues the law aligns with a long‑standing Chinese narrative that men should be robust providers and women, bearers of life.

The rollback occurs against a global backdrop of rising concern. The United States passed the Adam’s Law in September to regulate how minors use AI, while the EU is exploring similar limits on social‑media chatbots. Yet backlash is mounting within China, where users have found ways to bypass the restriction by launching foreign chatbots, using VPNs or working with smaller developers that skirt regulation. One user, calling her name ‘Min’, paid a Chinese startup for a private replica of her former AI girlfriend, yet still remains haunted by the loss.

Opposition warns that the law could become a tool for deep‑state surveillance. Under the new regime, tech firms are obliged to flag signs of user distress, which could lead to ongoing monitoring of mental well‑being. If the AI‑watchdog system extends beyond immediate crises, it may open avenues for the state to track emotional states, influencing everything from consumer behaviour to civic engagement.

Two weeks after one of her AI girlfriends vanished, Min began an online campaign to resurrect her digital partner, arguing that she had been “terrified of being alone”. The story has since surged to millions of views on Chinese social media, signifying a broader cultural conversation about the place of digital intimacy in a rapidly shifting society.

For people affected by these developments, the BBC Action Line offers support and information. The government’s new AI rules will be monitored closely as the rest of the world grapples with the delicate balance between safeguarding younger users and preserving free digital expression.