《人工智能安全治理的互操作性：伦理、法规与标准》报告在2025年联合国大学驻澳门研究所人工智能大会发布- 大数跨境

首页

《人工智能安全治理的互操作性：伦理、法规与标准》报告在2025年联合国大学驻澳门研究所人工智能大会发布

AIII人工智能国际研究院

2026-02-03

导读：Policy Report on Interoperability in AI Safety Governance: Ethics, Regulations, and Standards

Policy Report on Interoperability in AI Safety Governance: Ethics, Regulations, and Standards

政策报告：《人工智能安全治理的互操作性：伦理、法规与标准》

UNU Macau launched a new policy report proposing actionable pathways to bridge global AI safety governance

联合国大学驻澳门研究所发布新政策报告，提出弥合全球人工智能安全治理鸿沟的可操作路径

The United Nations University Institute in Macau has released a new policy report, Interoperability in AI Safety Governance: Ethics, Regulations, and Standards, offering critical recommendations to strengthen global AI safety frameworks through enhanced interoperability.

联合国大学驻澳门研究所发布了一份新的政策报告——《人工智能安全治理的互操作性：伦理、法规与标准》，提出了通过增强互操作性来加强全球人工智能安全框架的关键建议。

Drawing on country studies from China, South Korea, Singapore, and the United Kingdom, the report identifies effective tools and key barriers to interoperability in AI safety governance. It highlights practical ways to build a governance ecosystem that is globally informed yet locally grounded.

报告基于对中国、韩国、新加坡和英国的国别研究，识别了人工智能安全治理中实现互操作性的有效工具和主要障碍。报告着重阐述了如何构建一个既具全球视野又立足本土的治理生态系统。

Interoperability is a cornerstone of AI governance—essential for reducing risks, fostering innovation, enhancing competitiveness, promoting standardization, and building public trust. Yet progress remains hindered by fragmented regulations, limited global coordination, and insufficient engagement from the Global South.

互操作性是人工智能治理的基石——对于降低风险、促进创新、增强竞争力、推动标准化以及建立公众信任至关重要。然而，法规碎片化、全球协调有限以及全球南方参与不足等问题仍然阻碍着进展。

Focusing on autonomous vehicles, education, and cross‑border data flows, the report compares ethical, legal, and technical frameworks across the four countries. It identifies areas of convergence and divergence, offering policy recommendations aligned with the Global Digital Compact and relevant UN resolutions.

报告聚焦自动驾驶、教育和跨境数据流动三个领域，比较了上述四个国家在伦理、法律和技术框架方面的异同。报告识别了趋同与分歧领域，并提出了与《全球数字契约》及相关联合国决议相一致的政策建议。

The study adopts a regulatory learning approach, engaging stakeholders in each jurisdiction to capture local insights and foster interjurisdictional learning. Coordinated by UNU, the initiative introduces a collaborative model for transnational policymaking by identifying shared challenges and co‑developing strategic responses.

本研究采用了监管学习的方法，吸纳各管辖区域的利益相关者参与，以获取本地见解并促进跨管辖区域学习。该倡议由联合国大学协调，通过识别共同挑战和共同制定战略应对措施，引入了一种跨国政策制定的协作模式。

Recommendations span three dimensions—ethical, regulatory, and technical interoperability. They include advancing a global AI ethics framework, establishing multilateral systems for coordinated governance, enhancing transparency and accountability, promoting interoperable digital public infrastructure, and embedding interoperability by design in technical standards.

建议涵盖三个维度——伦理、监管和技术互操作性。具体包括：推进全球人工智能伦理框架；建立多边协调治理体系；增强透明度和问责制；促进可互操作的数字公共基础设施；以及将"设计即考虑互操作性"原则嵌入技术标准。

The report concludes that the future of AI safety governance is evolving toward evidence‑based, outcomes‑oriented models that complement principle‑led frameworks. Sustaining momentum will require policymakers to deepen normative specificity, enhance legal interoperability, expand interoperable standards, strengthen data governance, and invest in AI literacy and capacity building.

报告总结认为，人工智能安全治理的未来正朝着以证据为基础、以结果为导向的模式发展，这些模式将与以原则为主导的框架形成互补。要保持发展势头，政策制定者需要深化规范的具体性、增强法律的互操作性、扩展可互操作的标准、加强数据治理，并投资于人工智能素养和能力建设。

The study was a joint effort by multiple UNU Global AI Network members Yik Chan Chin, David A Raho, Hag-Min Kim, Chunli Bi, James Ong, Jingbo Huang, Serge Stinckwich, sponsored by SenseTime.

本研究由联合国大学全球人工智能网络的多位成员——Yik Chan Chin、David A Raho、Hag-Min Kim、Chunli Bi、James Ong（翁家良）、Jingbo Huang（黄京波）、Serge Stinckwich ——共同完成，并获商汤科技赞助。

Download the Full Report 下载完整报告

Method1 - Policy Report: Interoperability in AI Safety Governance: Ethics, Regulations, and Standards – Full Report PDF download link below:

方式一政策报告：《人工智能安全治理的互操作性：伦理、法规与标准》——完整报告 PDF下载链接如下:

https://collections.unu.edu/eserv/UNU:10363/Interoperability_in_AI_Safety_Governance.pdf

Method 2 - Download the full report through the following link:

方式二通过以下链接下载完整报告：

https://arxiv.org/abs/2601.06153

UNU Macau 网站报告发布链接

https://unu.edu/macau/news/new-policy-report-interoperability-ai-safety-governance-ethics-regulations-and-standards

About UNU Macau AI Conference 2025

In celebration of the United Nations' 80th Anniversary (#UN80) and the United Nations University’s 50th Anniversary (#UNUat50), the UNU Macau AI Conference 2025 successfully concluded on October 24 at the Galaxy International Convention Center in Macau. The event attracted over 500 participants from more than 30 countries, who engaged in in-depth discussions on the global development direction of artificial intelligence under the theme "AI for Humanity: Building an Equitable Digital Future".

为庆祝联合国成立80周年（#UN80）以及联合国大学成立50周年（#UNUat50），“2025年联合国大学驻澳门研究所人工智能大会”于10月24日在澳门银河国际会议中心圆满落幕，大会吸引了来自30多个国家的逾500位参与者，围绕“惠及全人类的人工智能：构建公平的数字未来”主题，深入探讨人工智能的全球发展方向。

UNU Macau大会网址 https://unu.edu/macau/aiconf2025

Professor Tshilidzi Marwala 教授开幕式致辞

In his opening address, Professor Tshilidzi Marwala, Rector of the United Nations University and UN Under-Secretary-General, emphasized the necessity of establishing an inclusive governance framework to advance technological progress while safeguarding humanistic values: "Three activities together form an integrated platform, promoting the exchange of leading global ideas, passing knowledge to the younger generation, and driving AI-powered sustainable development solutions to address challenges in areas such as climate resilience, digital inclusion, governance, and education."

联合国大学校长兼联合国副秘书长Tshilidzi Marwala教授在开幕致辞中强调了建立包容性治理框架的必要性，以在保障人文关怀的同时推动技术进步：“三项活动共同构建一个有机平台，促进全球领先的理念交流、知识传承予年轻一代，并推动以人工智能为驱动的可持续发展解决方案，以应对气候韧性、数字包容、治理及教育等领域所面临的挑战。

Dr. Huang Jingbo黄京波博士开幕式致辞

Dr. Huang Jingbo, Director of the United Nations University Institute in Macau, noted that the conference coincided with significant anniversaries for both the United Nations and the United Nations University. He stressed the importance of leveraging AI as a bond to strengthen global cooperation and build bridges toward a fair digital future that "benefits all humanity."

Dr. Huang Jingbo explained: "The event featured over 30 parallel sessions and five keynote speeches, exploring the application of AI in fields such as healthcare, education, youth development, employment, and climate change. Participants listened to diverse perspectives from academia, the private sector, governments, and other stakeholders."

联合国大学澳门研究所所长黄京波博士在发言中提出，本次大会恰逢联合国与联合国大学重要周年纪念，更需以AI为纽带强化全球合作，为“惠及全人类”的公平数字未来搭建桥梁。

黄京波博士介绍：“这次活动设30余场平行会议及5场主题演讲，探讨人工智能在医疗、教育、青年发展、就业及气候变迁等领域的应用，与会者聆听了来自学术界、私营机构、各国政府等多元声音。”