What is digital public infrastructure (DPI)?

Just as roads and water systems are built with public money, the idea is to build the basic parts of the internet, such as identity checks, payments and data exchange, as public systems that no single company owns. The book *Plurality* also calls for everyone to have access to information that has not been tampered with, as a right.

Roads, water systems and libraries are built with everyone's taxes, and anyone can use them. Online, though, we rely heavily on systems owned by large private companies, for logging in and for paying. "Digital public infrastructure" (DPI) is the idea of building these foundations of the internet as public things too. "Infrastructure" means the base that supports a society from below.

In short

The idea is to build the "basic parts" of the internet, such as identity checks, payments and data exchange, as public systems that no single company owns [1].

The book ⿻ Plurality goes further and calls for everyone to be able to access them in a way that delivers information that has not been tampered with, as a right. The book calls this "access with integrity" [2].

For example

India Stack. The book names the "India Stack", developed in India over the last decade or so, as the most prominent example of digital public infrastructure [3]. Part of it, Aadhaar, an identity system based on biometrics such as fingerprints, has enrolled more than 99% of India's population. It has become a key to social services, including a monthly ration that over 800 million people receive. MOSIP, an open-source identity platform (one whose design is public and which anyone can use and improve) inspired by the Aadhaar model, has been adopted in countries including the Philippines, Sri Lanka, Uganda, Morocco and Ethiopia [4].

The book says that the core of how the India Stack became so central to the private sector as well is that the government could shape the direction of the whole digital ecosystem through the standards it adopted and what it bought [5].

Ukraine's Diia and Estonia's mRiik. Both are public apps that give citizens secure access to government services and up-to-date information. The book writes that "Diia has shown the world how digital technology can break down long-standing corruption." Estonia launched mRiik, inspired by Diia, and Ukraine shared its code and design methods [6].

IPFS and public code in Taiwan. Taiwan is one of the world's main targets of attacks that flood servers with traffic (DDoS attacks). Its government adopted IPFS, which stores files spread across many computers, for its websites, making them more resistant to such attacks [7]. In a 2023 speech, Audrey Tang, then Taiwan's Minister of Digital Affairs, explained that the website of Taiwan's Ministry of Digital Affairs looks very much like the UK government's site because it embraces "public code". What democracies share is not just a design system but also "the cybersecurity audited forms and authentication and so on" [8].

In more detail

DPI as the book uses the term

Chapter 7-0 of the book writes that a pattern is spreading: a public mission to create infrastructure, with "participation but not dominance from the private sector", working with civil society so that new ideas can emerge in many places. This is "often labeled 'digital public infrastructure' (DPI)", the book says [1].

As examples it names the India Stack, the EU's European Digital Identity and the Gaia-X data framework, and experiments in Brazil (the instant payment system Pix) and Singapore (the instant payment system FAST). The book writes: "To a large extent, we are primarily advocating for this approach to be scaled up and become the central approach to the development of global ⿻ society" [1]. In 2023, the G20 digital ministers identified DPI as a key focus for worldwide cooperation [9].

On what scale?

The book pictures a goal for a decade from now: "we imagine that governments and charities will ensure we devote roughly 1% of GDP to digital public research, development, protocols, and infrastructure, amounting to nearly a trillion US dollars a year globally or roughly half of current global investment in information technology." That would "increase public investment by at least two orders of magnitude", that is, at least a hundredfold [10].

The actors would not be a single country, the book writes, but a network linking national digital ministries and open-source communities that anyone can join, across borders. As examples, the book names Taiwan's digital minister (appointed in 2016) and its Ministry of Digital Affairs (2022), Ukraine's Ministry of Digital Transformation, and Japan's Digital Agency, founded in 2021 and inspired by discussions with Taiwan [11].

"Access with integrity"

Chapter 4-5 says that the rights of digital life described in the earlier chapters mean nothing unless everyone can access that world securely and faithfully [2].

What it asks for is not mere access. "If the information some receive is accurate and others corrupted, it is worse than if the latter had no access at all," the book writes [2]. Following a line from the Universal Declaration of Human Rights of 1948 to a 2023 declaration addressing mass manipulation by generative AI, it holds that "we must ensure that everyone has equal access to contextually complete information" [2]. → contextual integrity

The divide, and two directions

According to the book, before the pandemic 76% of urban households around the world had home internet access, compared with 39% in rural areas: nearly double. The International Telecommunication Union (ITU) reports that in 2020 alone, 466 million people used the internet for the first time [12].

Taiwan, Estonia and the Scandinavian countries pioneered the principle of digital access as a foundational right. But the book says "this is just a downpayment on the long-term investment we need to support digital public infrastructure." It asks policymakers for two directions [13]:

Infrastructure linked across countries, to overcome problems that countries cannot solve without joining forces.
Dealing with realistic fake videos and voices (deepfakes), so that the meaning of information is protected.

The book cites the forestry expert Suzanne Simard's research showing how trees in a forest share warnings and nutrients through underground fungal networks. Digital infrastructure, too, should face threats together as an overlapping web of open standards, open-source software and open data, the book says [14].

What is debated

DPI identity tends toward centralization (the book's own criticism). Chapter 4-1 writes that the most influential developments in identity tools have taken place in developing countries, "often under the banner of 'Digital Public Infrastructure'". But the book criticizes many of them as taking a unified, centralized form based on biometrics [4]. On Aadhaar, it mentions the worry that those running the system gain the ability to watch citizens' activities across many domains (a concern activists have raised about the Muslim minority in India), and that India's Supreme Court limited its use by the private sector [4][15]. In other words, while the book promotes DPI, it argues that its identity components should be plural identity. → Identity and Personhood
The lack of legal vehicles. The book points out that there are few legal forms that let open-source nonprofits operate internationally. As an example, it notes that the Open Collective Foundation, which helped support the making of the book, was in the process of dissolving at the time of writing, and argues that legal structures are needed to support democratic nonprofits that work across borders [16].

Common misunderstandings

DPI means the state builds everything. DPI in the book's sense is something the private sector can take part in but not dominate, built together with civil society and open-source communities. The book rejects making the state the one single center [1][11].

Where it fits in Plurality

This corresponds to chapter 4-5, the last chapter of Part 4 of the book, "Freedom", which is framed by the rights operating system. The book's chapter summary says: "For all these to be true human rights, there needs to be universal access to these above capabilities with informational integrity" [17].

In the final policy chapter, 7-0, DPI becomes the center of the book's policy proposals. It is described as reworking, for today's world, the approach of the US research agency described in chapter 3-3, "The Lost Dao", and the experience of Taiwan's digital democracy [1][11].

The individual components are discussed in the chapters on Identity and Personhood (identity), Commerce and Trust (instant payments) and Property and Contract (IPFS and data sharing).

In 2024, Tang said that the idea of public digital infrastructure is not new to either Japan or Taiwan, but that it was mostly realized at the municipal level. International DPI was not recognized as a must-have until the pandemic, she said, when infrastructure that everyone needs, such as digital vaccine certificates, appeared [18].

Sources

Plurality, 7-0, end of the section "Lessons from the past" — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 4-5 "Access", introduction — https://www.plurality.net/v/chapters/4-5/eng/ ［CC0］
Plurality, 4-0, section "⿻ freedom" — https://www.plurality.net/v/chapters/4-0/eng/ ［CC0］
Plurality, 4-1, section "Public and decentralized digital identity" — https://www.plurality.net/v/chapters/4-1/eng/ ［CC0］
Plurality, 7-0, start of the section "⿻ regulation" — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 4-5, section "Infrastructures for information integrity", second half — https://www.plurality.net/v/chapters/4-5/eng/ ［CC0］
Same (the paragraph on IPFS) — https://www.plurality.net/v/chapters/4-5/eng/ ［CC0］
Audrey Tang, "Speech at LSE IDEAS" (2023-12-01) — https://archive.tw/2023-12-01-speech-at-lse-ideas ［CC0］
Plurality, 7-0, section "A new ⿻ order" (the paragraph on the G20) — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 7-0, opening vision of the future — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 7-0, section "A new ⿻ order" (the paragraph on digital ministries) — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, 4-5, section "Bridging the digital divide", paragraph 2 — https://www.plurality.net/v/chapters/4-5/eng/ ［CC0］
Same, paragraphs 1 and 3–5 — https://www.plurality.net/v/chapters/4-5/eng/ ［CC0］
Plurality, 4-5, start of the section "Infrastructures for information integrity" — https://www.plurality.net/v/chapters/4-5/eng/ ［CC0］
Plurality, 4-1, section "Public and decentralized digital identity", paragraph 6 — https://www.plurality.net/v/chapters/4-1/eng/ ［CC0］
Plurality, 7-0, section "⿻ regulation", paragraph 2 — https://www.plurality.net/v/chapters/7-0/eng/ ［CC0］
Plurality, chapter summaries (Chapter 4-5) — https://github.com/pluralitybook/plurality/blob/86158859464aee75633acd854c656928121a7fd8/contents/english/8-blurb.md ［CC0］
Audrey Tang, "Conversation with Hal Seki and Japanese delegates" (2024-03-22) — https://github.com/audreyt/transcript/blob/main/2024-03-22-Conversation-with-Hal-Seki-and-Japanese-delegates.md ［CC0］

We welcome error reports and questions ("What is…?"). Submissions become public issues.

Report an error in this article ・ Ask a question