In recent years, China's aspiration for global technology leadership has driven its significant investments in artificial intelligence (AI) for national security, economic growth, and societal well-being. Although there is increasing research and analysis on China's domestic AI development ecosystems and drivers, the details of China's development-financed AI exports remain elusive. Despite being the single-largest provider of foreign development assistance, Beijing does not participate in aid or debt transparency initiatives. To address this gap, researchers from the RAND Corporation and AidData jointly built a new database on China's AI export projects that are funded with official development financing: China's AI Exports Database (CAIED). CAIED uses data from multiple public databases and indexes related to China's global financing and recipient countries' electoral democracy, freedom, and data protection and privacy status. In this report, the authors analyze this quantitative dataset — adding qualitative country case studies based on interviews and social media analysis — to examine the distribution, technology, financing, and data safety aspects of China's AI exports.

With average annual commitments reaching $85 billion, the Chinese government is now the world's largest provider of development financing. Supported by large funding, many Chinese technology companies can deploy state-of-the-art artificial intelligence (AI) tools in development contracts in recipient countries. The AI exports facilitated by these arrangements are likely to bolster China's growing global AI technology-related supply chains, trade flows, technology standards, and regulatory systems. This report is the technical documentation for the China's AI Exports Database (CAIED), which is a tool that tracks Chinese government–supported development finance projects that used or enabled AI technology in the Global South between 2000 and 2017. The goal of the report is to explain the motivation, data source, and methodology of building the tool. This report also presents a detailed codebook for the database and explanations on how to use the interactive world map and country selector. CAIED was built on AidData's Global Chinese Development Finance (GCDF) Dataset version 2.0. Using advanced data mining tools, the authors identified 155 projects enabling AI exports and described their characteristics in rich detail. Users can filter the world map by AI technology categories associated with the development finance projects and learn further details of the exported China AI project by clicking on the country of interest.