During the general debate of the United Nations General Assembly, a high-level meeting on international cooperation for capacity-building in artificial intelligence was jointly initiated by China and Zambia. Foreign Minister Wang Yi introduced the “Artificial Intelligence Capacity-Building Action Plan For Good and For All” proposed by China.
The background of the “Action Plan” stems from the UN General Assembly resolution, “Enhancing International Cooperation on Capacity-Building of Artificial Intelligence”, i.e. A/RES/78/311. which China sponsored in July, with participation from over 140 countries. This program further advances the resolution’s specific measures based on the AI capacity-building seminar held in early September in Shanghai, China, with participation of UN Secretary General and UN experts. It represents an important progress in discussions on AI capacity-building on the UN platform. Previously, during the Summit of the Future, member states collectively adopted the “Global Digital Compact,” which includes an international consensus on advancing AI capacity-building through international partnerships. Following this, the high-level meeting on international cooperation for AI capacity-building convened at the UN, supporting and responding to the “Global Digital Compact” and contributing the first concrete action to its AI capacity-building initiative. Wang Yi’s proposal to establish a “Group of Friends for International Cooperation on AI Capacity-Building” will be a key component in advancing the AI capacity-building initiative of the Global Digital Compact at the UN.
The “Action Plan” articulates China’s systematic understanding of AI capacity-building through its vision and goals, encompassing infrastructure, industry and ecosystem empowerment, talent development, public literacy enhancement, data construction, and safety governance. Advancing digital and AI hardware and software infrastructure is fundamental and essential for capacity-building, and this is precisely what many low- and middle-income countries currently lack. Only by establishing a solid foundation can these countries effectively leverage AI to empower various industries according to local conditions. Safety and governance are vital components of AI capacity-building, ensuring the robust development and application of AI. Improving public literacy allows a nation to have an objective understanding of the developmental stage and level of AI, enabling appropriate collective decisions based on genuine development realities. Teaching people to fish, rather than just giving them fish, emphasizes that every country has the sovereignty to cultivate its own talent in developing original AI capabilities, making talent development crucial for sustainable capacity-building.
To achieve these visions, the “Action Plan” outlines ten specific actions from China. These actions focus on collaboratively building data and service infrastructures inclusively, establishing joint laboratories to ensure that capacity-building is grounded on solid foundations. The program advocates for international cooperation in the form of open-source communities, fostering collaboration in application ecosystems. This approach allows countries with advanced AI technologies and services to assist underdeveloped countries and regions in bridging the digital divide in the most transparent and direct way. Data and corpora for AI training must come from local sources to incorporate local civilisations and cultures into the AI learning and training process, thus better serving specific regions. Regarding AI ethics and safety, developed countries are encouraged to actively share their experiences in evaluation and regulation, utilizing interoperability for shared safety and development.
The “Action Plan” represents a consensus and shared understanding between the Chinese government and the AI academia-industry-research community, providing strong support, participation, and practice towards the UN’s vision for AI capacity-building. It calls for active responses and participation from various sectors in China to jointly advance AI capacity-building with partners from different countries, empowering deep cooperation in global AI development and governance.
Author:
Zeng Yi, Professor at the Institute of Automation, Chinese Academy of Sciences, Director of Beijing Institute of AI Safety and Governance, Member of the UN High-Level Advisory Body on Artificial Intelligence.