BANNER: Cover imagery derived from the Catalan Atlas (1375) by Abraham Cresques, held in the Bibliothèque nationale de France. Digital scan sourced from Wikimedia Commons. Fiber optic cable routes representing real Digital Silk Road infrastructure (CPEC, PEACE, AAE-1, TASIM) were composited using Python and the Pillow imaging library

China’s Digital Silk Road (DSR)—the digital economy pillar of the Belt and Road Initiative—has become a central channel through which Beijing engages Global South countries on digital infrastructure and, increasingly, artificial intelligence (AI). Initially focused on “hard” connectivity (for example, telecom networks, fiber and submarine cables, etc.), since 2023, the DSR has evolved toward “soft connectivity”—encompassing AI governance diplomacy, standards-setting, capacity-building, and service-layer platforms. This shift reflects both China’s domestic economic imperatives (in particular, high-tech overcapacity and constrained access to advanced markets) and its geopolitical objective of shaping a state-centric alternative digital order that elevates government control over rights-based governance.

The DSR operates through a hybrid governance model described by the People’s Republic of China (PRC) as “government guidance and enterprise leadership.” In practice, it is best understood as a flexible framework rather than a single top-down program: central guidance sets broad priorities, while firms and subnational actors shape implementation, sometimes retrofitting existing commercial projects into DSR branding. This market-led structure amplifies the role of Chinese firms, which translate policy objectives into durable presence through competitive pricing, integrated offerings, and promotion of technical standards.

A defining feature of the DSR’s operational model is integrated deployment via a “Chinese tech stack.” Firms increasingly bundle telecom networks, cloud/data centers, applications, and governance frameworks. Adoption at one layer increases incentives to adopt adjacent layers for interoperability and operational continuity, raising switching costs and creating technical and institutional lock-in. This approach is reinforced by coordinated dealmaking that links firms with policy banks and quasi-state intermediaries (including, for example, overseas Chinese chambers of commerce) to reduce transaction costs, manage risk, and accelerate uptake. China’s open-source AI development and promotion and localized offerings (including expanded provision of local language services, sovereignty-conscious data center and cloud offerings, and “AI-in-a-box” solutions optimized for constrained compute environments) further strengthen the stack’s appeal in Global South contexts.

The DSR’s influence is reinforced through multilateral coalition-building and agenda-setting in AI governance. China has been particularly effective in United Nations (UN)-centered processes and South–South groupings, framing AI governance in terms of capacity-building, development, and sovereignty-related priorities. Alignment is visible among key partner states (including Saudi Arabia, Ethiopia, Nigeria, Vietnam, and Cuba, among others), in coalitions such as the BRICS economic grouping and the Group of 77, and in UN bodies and expert mechanisms where PRC narratives gain legitimacy by consciously reflecting developing country priorities.

At the same time, systemic weaknesses complicate the durability of China’s approach. These include persistent concerns around privacy, surveillance exports, and lack of oversight; debt sustainability and opaque collateralization practices; growing awareness of the risks and downsides of long-term technological dependency; geopolitical backlash and trust deficits amid intensifying US-China competition; and socio-cultural challenges adapting solutions to local contexts, compounded by uneven ESG (environmental, social, and governance) performance. These frictions create opportunities for alternative partnership models where governments seek greater transparency, resilience, and genuine local ownership.

For like-minded democracies, the strategic challenge is not to “counter” China forum by forum, but to compete by offering a differentiated model built around partner agency, interoperability, and long-term sustainability. Effective engagement must be grounded in concrete offers, encompassing financing, infrastructure, and capacity-building, and should be framed around the priorities of partner governments. The report maps a set of recommended venues for democratic coordination on messaging and strategy and flags arenas where strategic capture or agenda distortion is more at risk, requiring tighter collective coordination around strategy and messaging.

The report outlines specific, actionable recommendations for like-minded countries in collaborating with Global South partners on AI.