Data sharing is increasingly becoming foundational to how states coordinate sectors, deliver public services, and integrate AI – particularly in low- and-middle-income countries. As governments move beyond digital identity and payments, data sharing is emerging as the critical third pillar of Digital Public Infrastructure (DPI). How these systems are designed is no longer a technical question alone, but one of public interest, equity, and long-term trust. As a step towards our vision of building better data-sharing systems as DPI, we are launching our foundational report on data sharing as DPI. The report is a part of our larger work with the Gates Foundation on trusted data sharing.

Data sharing as DPI holds enormous potential for public good by enabling collaboration across institutional, sectoral and technical boundaries. This includes sector coordination, market-led innovation, compliance, the trustworthiness and authenticity of claims in multi-stakeholder ecosystems, the coordinated delivery of integrated services with resilient and sovereign systems, and the stewardship of data as a public asset, with state system efficiency and modernisation.

Data sharing systems, however, are only as strong as their foundations. Open data flows without a strong governance foundation can undermine trust, exacerbate exclusion, and ultimately weaken the very systems they aim to improve. Data sharing that prioritises people recognises that the value of data needs to be realised in specific contexts and is stakeholder-dependent. Open flows of data must therefore be underpinned by ownership, accountability and equitable control. Most importantly, governance, that is people-first, determines long-term sustainability – not just technology.

Our work adopts a socio-technical approach, examining how data-sharing systems are built, and how they interact with state capacity, political priorities, and societal contexts. The analysis draws on a review of over 120 pieces of global literature, a comparative study of about 50 data-sharing models across nine sectors, and interviews with experts spanning government, technology, policy, and development practice.

Across contexts, we found that state priorities and institutional capacity fundamentally shape design choices in data-sharing models. The current governance and enabling ecosystems are fragmented, and a technology-first approach may not adequately address pertinent issues like meaningful ownership of data and diverse context-specific infrastructure needs. Fundamentally, trust underpins all effective data-sharing systems, and across contexts, a governance-first approach emerged as essential.

As data sharing becomes central to DPI and even AI ecosystems, the choices made to build data sharing systems can shape inclusion, resilience and public trust in all contexts. Aligned with the broader priorities of this work, this report aims to provide a foundation, exploring key considerations on data sharing from literature, lessons from existing models and highlighting the need for a taxonomy on data sharing. In the following months, this report will be complemented with a comprehensive taxonomy, a use-case repository and a self-assessment tool to support governments and ecosystem actors in making informed, context-sensitive design decisions.

You can read the report here:

Download