Recently, the Cyberspace Administration of China has unveiled the "Interim Measures for the Administration of Human-like Interactive AI Services (Draft for Public Comment)", seeking input from a wide range of stakeholders. The Draft explicitly mandates that providers of human-like interactive AI services must clearly and conspicuously inform users that their interaction is with artificial intelligence, rather than with a real person.

The measures define human-like interactive AI services as products or services that leverage AI technology to mimic human personality traits, thought processes, and communication styles. These services engage in emotional interactions with users via text, images, audio, video, and other multimedia formats.

Furthermore, the measures establish a guiding principle that balances healthy development with lawful governance. They encourage innovation in service offerings while implementing an inclusive, prudent, and tiered regulatory approach.

Regarding service standards, the measures outline several prohibited activities. They also require service providers to assume primary safety responsibilities, establish and refine relevant management systems, and ensure safety throughout the entire service lifecycle.

Clear guidelines are provided for data security and user protection. This includes the utilization of datasets that adhere to core socialist values, safeguarding user data security, and offering mechanisms for user status identification and intervention.

Additionally, the measures institute special protective measures for vulnerable groups such as minors and the elderly.