TLDR: China has introduced new draft measures aimed at strengthening ethical oversight in artificial intelligence research and development. Issued by a coalition of 11 government agencies and research institutes, these proposals build upon earlier regulations from 2022 and 2023, providing more specific and targeted review requirements for the rapidly evolving AI sector. The draft measures address potential ethical risks across various domains, including human health, dignity, environmental impact, and public order.
On August 26, 2025, China took a significant step towards refining its regulatory framework for artificial intelligence by unveiling new draft measures focused on AI ethics. These comprehensive proposals, jointly issued by 11 prominent government agencies and research institutes, signal Beijing’s commitment to ensuring responsible AI innovation and deployment.
The new draft measures are designed as a detailed follow-up to the ‘2022 Opinions’ and the ‘2023 Measures on sci-tech ethics,’ specifically targeting the unique challenges and ethical considerations within the AI sector. This initiative underscores a proactive approach to governance, moving beyond general scientific ethics to address the specialized complexities of AI.
The scope of these draft measures is broad, encompassing AI research and development activities within China that could pose ethical risks. These risks are identified across critical areas such as life and health, human dignity, environmental impact, public order, and sustainable development. The regulations also extend to any other AI-related activities mandated for review under existing national laws and regulations.
Also Read:
- China’s AI Future: 36Kr’s 2025 Industry Conference to Unveil “Chinese Solution”
- HBKU to Convene Global Conference on AI Ethics in September
This latest move aligns with China’s broader strategy to establish a robust regulatory environment for AI. Previous guidelines have emphasized several key provisions, including stringent ethical standards to ensure AI systems respect human rights and prevent discrimination, particularly in sensitive applications like healthcare, finance, and employment. Data security and privacy protections are also paramount, with requirements for robust safeguards, controlled access, and regular audits for AI systems handling sensitive information. Furthermore, transparency in AI decision-making processes is a core tenet, mandating documentation of how AI systems arrive at conclusions to foster accountability and user trust. Safety and risk assessment protocols are also integral, requiring regular evaluations of AI systems, especially those operating in high-stakes environments.
