Navigating AI Morality: PRC’s Vision for Ethical AI Governance

On 22 August 2025, the Ministry of Industry and Information Technology of China (“MIIT”) unveiled a consultation paper on the Measures for the Ethical Management Services of Artificial Intelligence Technology (for Trial Implementation) (“Draft Measures”). The Draft Measures signal a significant step towards formalising ethical oversight in AI development. Under the Draft Measures, certain AI research or development activities (as further set out below) would be subject to mandatory ethics reviews and ongoing compliance obligations – marking a shift towards a more structured and accountable innovation in the field of artificial intelligence.

Here are some key highlights of the Draft Measures:

Scope of the Draft Measures	The Draft Measures apply to higher education institutions, research institutions, medical and healthcare institutions and companies that engage in AI scientific research and technological development activities conducted within Mainland China which may pose AI ethical risks in areas such as life and health, human dignity, ecological environment, public order and sustainable development, and any other technology activities requiring AI ethics review as imposed under other PRC laws and regulations (“AI Technology Activities”).
Establishment of Ethics Committee & Service Centres	Entities engaging in AI Technology Activities, if feasible, should establish an AI Technology Ethics Committee (人工智能科技伦理委员会) (“Ethics Committee”) in order to oversee the AI ethical governance within the organisation if it is capable to do so. The Ethics Committee should be composed of experts in AI technology, applications, ethics, law and other related fields. Local regulatory authorities may, based on practical needs, collaborate with relevant organisations to establish a specialised AI Technology Ethics Service Centre (人工智能科技伦理服务中心) (“Service Centre”), in order to provide key support services (e.g. conducting ethics reviews, offering training and delivering expert consultation) for AI-related activities.
Requirements for ethics review	Persons responsible for AI Technology Activities must first submit a request for an ethics review by their organisation’s Ethics Committee (if applicable) or a Service Centre. Approval must first be obtained before such AI Technology Activities can be undertaken. There are a number of materials which must be submitted as part of the ethics review – e.g. details of the contemplated AI Technology Activities, assessments regarding technology ethical risks, prevention and control plans, etc.. The Ethics Committee or Service Centre shall assess an application based on certain qualitative thresholds (e.g. ensure fairness and justice, ensure controllable and trustworthy use of AI, etc..), and shall provide a decision within 30 days of accepting an application (subject to extension in the event of complex circumstances or the need to supplement or correct materials). An applicant may file an appeal within 3 working days after receiving a decision of disapproval or order for amendment in case of any disputes. In certain circumstances, e.g. where the likelihood and extent of ethical risks are no greater than routine risks encountered in everyday life, a simplified procedure may be applicable.
Requirements for expert ethics review	Persons undertaking certain AI Technology Activities are generally required to engage in a further round of expert review upon completion of the ethics review as set out above. This includes the activities that are set out in the list of “AI Technology Activities Requiring Technology Ethics Expert Review” (“需要开展科技伦理专家复核的人工智能科技活动”清单) (the “Expert Review List”). The Draft Measures provide that the following AI Technology Activities require further expert review: development of human-machine integration systems which have a strong impact on human behaviour, emotions and life or health; development of algorithmic models, applications, and systems which may mobilise public opinion and influence social awareness; and development of highly autonomous automatic decision-making systems intended for scenarios involving safety or health risks.
Ongoing obligations	Follow-up reviews shall be conducted on approved AI Technology Activities at least every 12 months (and at least every six months for activities subject to expert review). Any changes in ethical risk exposures shall be promptly identified and reported by responsible persons to the relevant Ethics Committee or Service Centre. Organisations will also need to complete registration and filings on the designated platform regarding their Ethics Committees and, if applicable, details of activities falling under the scope of the Expert Review List. All organisations engaging in AI Technology Activities must establish clear channels for reporting violations of AI technology ethics.

What’s next?

The Draft Measures are yet to be finalised - the consultation period will end on 22 September 2025.

That said, organisations that engage in these high-risk AI activities should be prepared to review its current governance structure against the Draft Measures. Failure to implement these measures may lead to penalties in accordance with the PRC Law on Progress of Science and Technology and other relevant laws and regulations.

If you would like to understand more on what this means to your organisation, please don’t hesitate to contact us for more information.

The materials on the Eversheds Sutherland website are for general information purposes only and do not constitute legal advice. While reasonable care is taken to ensure accuracy, the materials may not reflect the most current legal developments. Eversheds Sutherland disclaims liability for actions taken based on the materials. Always consult a qualified lawyer for specific legal matters. To view the full disclaimer, see our Terms and Conditions or Disclaimer section in the footer.