Draft Working Paper

About this document

This paper presents preliminary insights and analysis based on a comprehensive but non-systematic review and narrative synthesis of global developments around DPI. It draws on grey literature, media coverage, and direct reporting from those involved. We welcome comments and feedback.

Authors:

Dr Stephanie Diepeveen

Dr Sumedha Deshmukh

Anonymous**

Suggested citation: Diepeveen, S., Deshmukh, S., Anonymous (2025). “The Global DPI Agenda: Promises vs. Realities in the Evolution of DPI for Digital Transformation.” Draft working paper, December. Available at:

**This draft is the product of the collective reflections of the authors, some named and some anonymous, on their experiences and analyses of the global agenda for digital public infrastructure (DPI). Our intention is to expand the discursive space in which DPI is interrogated beyond showcasing its benefits. Due to constraints around the critical exploration of DPI's challenges and harms, some authors have chosen to remain anonymous.

How to read this document

This document is meant for academics, activists, and civil society organisations that have been observing and engaging with the ongoing discussion on digital public infrastructures. It assumes a familiarity with the arguments for digital public infrastructure. It aims to raise questions on the value and validity of this approach by offering arguments for why it must be challenged globally. The authors envision this document as a starting point for creating a new research agenda, convenings, and better-equipped engagement with the DPI approach anchored not just in the promise of DPI but rather its reality as experienced by many countries globally. In an ideal scenario, this can be a starting point for considering alternatives to the current dominant vision and method of DPI supply and scale. And, with this, provides for a more balanced and inclusive discussion of possibilities and harms of digital public infrastructure globally, especially across countries of the Global Majority.

We have starred (⭐) key overview sections in the Table of Contents below. We’ve also included an Index of Themes if there’s something specific you’re interested in.

In this endeavour, we offer some historical context on the evolution of DPI and its narrative, and provide a reading list to support the reading of this document, especially for those less familiar with the history of digital public infrastructure and the context for its emergence as a distinct area of policy and implementation.

Executive Summary

Digital Public Infrastructure (DPI) has become firmly embedded within development, not the least as a mechanism to achieve the realisation of the SDGs. The global ecosystem of DPI supply has successfully established DPIs as a “silver bullet” solution at a time where the world is in unprecedented conflict, and when development goals as defined in the sustainable development goals (SDGs) are far from being met.

In its foundations, DPI is bound up with India’s approach to digital transformation through digital ID, data exchange and digital payments, though it has evolved and expanded to refer to a much wider set of government-led digital transformation efforts. As we explore in this paper, this definitional expansion appears in part intentional, to package the DPI approach for global audiences and distance itself from India’s checkered past with the development and deployment of public technologies.

As the DPI agenda grows, both the products that encompass the “DPI” and the processes that drive its global export need to be examined carefully to understand whether the promise of the technologies are being met. Through the supply-side push of DPI globally, there are growing concerns about the pace of DPI development and deployment, and whether there are reasonable and necessary frictions that take into account the situational analysis of where these technologies are being exported to. The narrative for support of DPI presents these tools as efficient, effective and economical for countries of different sizes and capability.However, there has been very little systematic attention to the realities of DPI in practice, across countries, and the strength of the evidence to support claims about implementation and outcomes.

Despite the growing optimism around DPI as an approach to digital transformation globally - from small island states to the EU - very little about the actual and potential harms, and how they might be mitigated. Published evidence that engages with real and potential challenges remains limited, and, early on, was written off as teething troubles. Therefore, it is critical to engage with the evolution of DPIs in an all-encompassing way to evaluate whether DPIs have actually delivered on stated goals, and its impact on vulnerable groups such as migrants/refugees especially critical in the context of Europe right now. Reflecting on the discourse around DPI in Europe, a concern emerges that the cast of characters, and the core ideology driving DPIs is the same - and now with global attention and interest.

<aside>

This working paper aims to broaden the foundations for research and discussion about DPI by first, mapping out the challenges around DPI implementation and outcomes, and two, proposing a framework through which to interrogate existing and potential harms.

Four critical areas of harm:

Harm to ordinary citizens, as they likely are paying for DPI in ways they do not fully understand (both now and in the future) as well as eroding channels for democratic accountability
**Exclusionary and uneven forms of private sector involvement,** especially in smaller and lower-income countries where DPI is being pushed
A loss of state capacity and democratic accountability, given dependencies within the design and implementation of DPI and flows of data while not necessarily delivering on objectives of enabling competition and sovereignty, and
Reinforcing of particular state capacities as a more authoritarian entity with new potential to exercise targeted harm, as people are recognised, interact and can be acted upon through integrated digital systems

Key features of how the global DPI agenda has (intentionally and unintentionally) been pushed have contributed to these four areas of harm. Some of the key dynamics raised in this paper include:

Market-enablement and market-making: DPIs enable private sector (technology companies and capital) to enter into markets that they traditionally would not be players in and create services/protocols that the private sector can use for innovation. Measurements of success, then, have also been closely tied to the private sector activity enabled by DPI.
Commercialisation and threats to competition: DPI is being implemented in ways that can limit competition within the wider ‘stack’. It opens public functions to private actors, interweaving commercial, personal and political interests in ways that can challenge democratic accountability but are not discussed.
Under and mis-use of data: DPI transactions generate a significant amount of data, but the discourse on DPI does not discuss the value of data, and the claims of the “public” on this data. This becomes increasingly important with discussions of ‘DPI for AI’.
Failures of sovereignty: How the DPI agenda is unfolding, especially with public services on private rails, bring complicated risks to sovereignty. This has included a focus on sovereignty from a few actors and not others. This also obscures risks to long-term state capacity potentially generated through public dependencies on private actors, as well as vulnerabilities to state capacity linked to wider supply chains. Compounding this, it is unclear how DPI models will be sustained in terms of funding and state capacity.
Inattention to inequality and distributional issues: The DPI agenda is being promoted in ways that fail to challenge inequality or distributional issues, meaning it risks further excluding already-excluded and/or marginalised people.
Failures of openness and ‘open-washing’: Openness is positioned as the silver bullet for innovation and competition. This can, in turn, obscure challenges associated with business models and sustainability of projects, as well as operational considerations that may lead to rent seeking or lock-in. </aside>

These dynamics are underpinned and sustained through particular narratives, a limited evidence base and concentrated funding sources. Narratives about DPI are sufficiently ambiguous to refer a wide range of offerings, while at the same time equating DPI with intended outcomes. The evidence base is thin at best, and, at worst, includes unsubstantiated and unjustified numbers. Funding is concentrated among a few actors, resulting in disproportionate influence over a wide range of initiatives.

As the DPI discourse is pushed in Europe and other parts of the world and emerges as a reasonable mechanism for the development of AI technologies, it is imperative to reckon with its problematic past in India, and track record in parts of the African continent. It is also important to conduct a transparent assessment of the supply side push and who benefits from it, so that developmental objectives can be delineated from agendas. DPIs are potent technologies which need to be governed to ensure that the promise of public interest and public scrutiny is delivered.

1. Introduction