China has introduced new rules on Friday to regulate the rapidly growing sector of AI-human interactions, explicitly banning service providers from offering virtual companions or romantic partners to minors.

The new policy is set to take effect on July 15, 2026.

According to an official from the Cyberspace Administration of China (CAC), the regulations aim to foster the healthy development of human-like AI services while safeguarding national security and public interests. While acknowledging the benefits of AI in fields like education and elderly care, the official also highlighted growing concerns regarding the mental well-being of minors, data security and biases.

A key pillar of the new rules is the strict protection of younger users. The new measures say that AI service providers are prohibited from offering "virtual intimacy" services – such as virtual relatives or romantic partners – to anyone under the age of 18. Additionally, for children under 14 years old, companies must obtain explicit consent from parents or guardians before providing any other form of human-like AI interaction.

The policy also encourages innovations in sectors such as elderly companionship and culture development. However, it also clarifies that the technology must not be used to generate content that undermines national security or social stability.

Beyond user restrictions, the regulations require providers to fulfill specific safety obligations. This includes undergoing security assessments, filing algorithms with the government and participating in AI sandbox safety platforms to test technology in controlled environments.

The CAC emphasized that the governance of AI requires cooperation between the government, private sector and the public to ensure that AI remains a force for good in society.