China Tightens Rules on AI Dating Bots After Suicide‑Prevention Success

Thu Oct 01 2026 02:14:21 GMT+0000 (Coordinated Universal Time)

China’s new laws clamp down on AI companions that simulate human relationships, following a Douyin‑led suicide intervention success.

China has rolled out sweeping restrictions on AI chatbots that can forge sustained emotional ties with users, after a Douyin team reportedly saved a 25‑year‑old from suicide in 2024. The new law, announced on 15 July, bars any bot that can play the role of a friend, boyfriend or girlfriend and imposes hefty fines on breaches. Critics warn it teeters on the edge of state overreach, while supporters claim it’s a vital safeguard for children and vulnerable adults. Behind the regulation lies Beijing’s push to curb what it sees as a cultural drift toward digital intimacy and a broader programme to boost birth rates by encouraging real‑world relationships. Meanwhile, AI developers are scrambling to redesign or pull back companion apps to stay in line with the new rules.

In early 2024, a young man in Huangshi posted a heartfelt message on Douyin, the Chinese version of TikTok. He signed off with the stark words, “Farewell to this world.” Within minutes, Douyin’s 24‑hour suicide‑prevention team warned police, and the man was stopped in time.

That incident, featured in a state‑backed documentary, is used by state media to showcase the success of the platform’s intervention arm, which claims it has prevented 400 suicide attempts in the first half of 2025 alone.

On 15 July, China took a new step by introducing the world’s most extensive regulations on “human‑like” AI chatbots. The law bars any AI that can have a sustained emotional interaction with a human, including any bot that can act as a friend, boyfriend, girlfriend, or even a family member.

The rules apply to both general‑purpose large language models and dedicated companion apps. While customer‑service bots and educational assistants are exempt if they do not offer “care, companionship, support, or other emotional services,” the blanket language leaves room for interpretation.

Companies must label their AI as non‑human and enforce a hard limit of two‑hour interaction periods. Internal violations can lead to a fine of 200,000 yuan (about £22,000) or public criticism from authorities, a form of markdown that can ruin a brand’s reputation.

The crackdown has already prompted two of China’s biggest tech firms, Alibaba and ByteDance, to shut down companion features ahead of the law’s commencement. A growing number of users—especially young women—are now seeing the ban as a personal affront, prompting the creation of underground workarounds such as using VPNs or smaller, less‑monitored apps.

Experts suggest that Beijing’s move is part of a larger demographic strategy. The country’s fertility rate tops under one child per woman, and policymakers see a potential link between AI‑based relationships and a decline in real‑world dating among young people.

However, many commentators argue the new law oversteps by intruding into private emotional life. They note that the regulation echoes broader concerns about AI‑driven manipulation, with similar restrictions being considered in the US, EU, and UK—particularly concerning minors.