Sovereignty

This piece is part of Reframing Impact, a collaboration between AI Now Institute, Aapti Institute, and The Maybe. In this series we bring together a wide network of advocates, builders, and thinkers from around the world to draw attention to the limitations of the current discourse around AI, and to forge the conversations we want to have.

In the run-up to the 2026 India AI Impact Summit, each piece addresses a field-defining topic in AI and governance. Composed of interview excerpts, the pieces are organized around a frame (analysis and critique of dominant narratives) and a reframe (provocations toward alternative, people-centered futures).

Rafael Grohmann is an assistant professor of media studies (critical platform studies) at the University of Toronto. His research focuses on digital labor, AI in the cultural sector, workers’ organizing, platform cooperativism, and the digital solidarity economy, especially in Latin America. 

In this conversation, Grohmann unpacks digital sovereignty. He historicizes the concept, tracing it to 1970s anti-imperialist political economy, which examined how global power and economic dependence shaped national development. Today, he argues, digital sovereignty has returned in a depoliticized form as the site of a battle between states and Big Tech that often sidesteps questions of who holds power and who benefits. In place of such top-down uses of the concept, Grohmann introduces a bottom-up notion of “popular digital sovereignty” based on his research with social movements and cooperatives in Latin America, which prioritizes community control, education, and the sharing of resources. A critical part of this approach is building and circulating prototypes of an alternative world.

Following is a lightly edited transcript of the conversation.

States are trying to use the idea of digital sovereignty against Big Tech, and Big Tech is co-opting it to sell it back to states.

The notion of digital sovereignty appeared in a time when digital platforms started to overcome state and civil society, especially after 2016 when Trump won the US election for the first time. In countries like Brazil or regions like the European Union, sovereignty became a way to address tech dependency and to address the way some platforms try to circumvent rules. The EU, especially, tried to bring strong regulation for platforms and for AI. 

In response, two or three years ago Meta, Alphabet, and Amazon launched digital-sovereignty programs. You can ask, “Is this a joke? Why is Amazon, the leader in infrastructure services and cloud services around the world, launching a digital-sovereignty program and what does that mean?” This is not something new. Platform companies are always trying to co-opt discourses to empty them of their political meanings and reappropriate them for themselves. They are reappropriating sovereignty as something individual, something you can claim, you can buy. So if you are a government, especially in the Global South, and you want to be sovereign but you don’t have the infrastructure or the money to be sovereign in a proper way, they can sell sovereignty to you. What that means is they install a local cloud in the country and say, “You are sovereign now because the cloud and the data centers are in your own territory, but they are owned by us.” In both a material sense and a discursive sense this is a way to sell sovereignty as a service.

Sovereignty’s resurgence is a symptom of the historical failure to reckon with colonialism.

Sovereignty is not a new word. Canada in the ’70s was talking about tech sovereignty. In that time, in countries in Latin America, there were Marxist theories of dependency. Being a dependent country in relation to others in the Global North in terms of technology means that you are not sovereign in that area. So sovereignty was one of the opposites of dependency.

Nowadays, we are living in a kind of Stranger Things nostalgia, reviving these old words like imperialism, colonialism and sovereignty because we didn’t address those issues properly in the past. The monsters and zombies are back because we are seeing imperialism and colonialism return in the worst form. Sovereignty has also come back with a renewed face.

The word “sovereignty”—and its close relative, “governance”—whitewash class struggle.

Another hype word related to “sovereignty” is “governance.” If I name my talk at the AI Summit “governance,” “sovereignty,” and “decolonial,” I will have the most attendance. Governance is sometimes framed as workers having a seat, government having a seat at the table. Everybody has a voice in this so-called governance. But really, it is class struggle. It’s not equal. It’s profoundly unequal and unfair, because what you call “AI governance” is someone having more voice and more participation than others. It’s a fake word to name real politics.

REFRAME: Popular Digital Sovereignty is a bottom-up, people-centric alternative.

Since 2018, the Brazilian Homeless Workers Movement’s technology sector has built people-centric alternative technologies.

I’ve been working with worker collectives and social movements for the past six or seven years, especially the Homeless Workers Movement [Movimento dos Trabalhadores Sem-Teto, or MTST] in Brazil. One of the strongest urban social movements in Latin America, MTST is struggling for housing and the right of housing in urban places.

In 2018, when the far right won the election in Brazil under the presidency of Jair Bolsonaro, the movement started to respond to how the institutional left was losing the battle in the digital environment. The far right was dominating the battles and ideologies on the internet and the institutional left didn’t understand the role of technology beyond being a tactical tool. So MTST decided to create a tech sector. 

They started to organize tech workers who were unhappy working for large tech companies in Brazil to create technology for homeless people. Now they have created a lot of these services. One of them is called Contrate Quem Luta [Hire Those Who Struggle]. It’s a kind of Taskrabbit governed by MTST that hires activists from the movement to be construction workers and domestic workers. So MTST activists go to clean your house and they also help to popularize the movement. 

This movement is inspired by the Brazilian educator Paulo Freire. Paulo Freire has a powerful statement from the ’60s or ’70s: “Reading the world precedes reading the word.” They paraphrased Freire, saying that “reading the world precedes reading the code.” So they started to teach computation to people in the suburbs of Brazil, especially in Sao Paulo, teaching people how to code without using a computer and from their own perspective. So it’s a movement about education, about circulation of struggles, and about imagination.

MTST calls this approach “popular digital sovereignty”—communities working on technology with, beyond, despite, and against the state.

The movement also started to theorize digital sovereignty from their own perspectives, which is a kind of epistemic sovereignty. They started to think that a social movement that struggles for housing needs to talk about the importance of creating alternative, worker-led frameworks for community-owned technologies. 

They started from the fact that people can’t imagine what an alternative looks like. So they adopted “digital sovereignty” as a guide and reframed it as “popular digital sovereignty.” In Latin America, “popular” is not a bad thing; it refers to something from people and from communities. There’s something powerful in that: Communities can control tech and infrastructure from below, can design technologies from their own perspectives beyond Big Tech, and they can control the infrastructure—sometimes with the state, sometimes beyond the state, sometimes despite the state, and sometimes against the state. 

The role of the state is important in this notion of sovereignty. But the way social movements are reclaiming digital sovereignty from below is a reminder that regulation and policies from the state are not enough. Ordinary people are putting pressure from their own perspectives and not waiting for policymakers. 

We need more cooperation between grassroots movements, sharing resources and approaches.

The groups trying to fight against Big Tech or creating alternatives are often a bit isolated. We need to create more cooperation among grassroots movements across the board, between tech workers, data workers, social movements, civil society, and coalitions in order to address AI or the technologies we want for the future. 

The cooperative response to digital economies is not each city creating their own app or their own AI version, but sharing infrastructures and projects. In the so-called gig economy, European delivery workers created an app shared by cooperatives in cities like Barcelona or Bordeaux. They are sharing the same app or sharing these infrastructures and sharing the costs for it in a federation called CoopCycle.

These nascent alternatives are “prototypes of struggles.”

These examples, from CoopCycle to the Homeless Workers Movement in Brazil, are what I call “prototypes of struggles.” They are prototypes because they are experiments, laboratories—which are called “laboratories” because they are made of so much labor. I don’t like it when academics, especially journalists, try to frame them as finished projects because they try to evaluate them and say they are failing. They are not startups. I call them prototypes of struggles not only because of class struggle, but also because they are struggling to survive and we need to create more coalitional power around them. 

There is a question of how we will learn about these alternative experiences or about these prototypes of struggles. Everybody knows Netflix but they don’t know that there is a worker-owned streaming service based in Detroit called Means TV. Initiatives like this one, MTST, or others suffer a lack of visibility in political spaces. So we have to circulate them.

We need to recover failed and invisibilized alternatives from the past to imagine and build possible futures.

For me, it’s a politics of memory as well. In the ’70s in Latin America, there was a lot of discussion about dependency, and there were also a lot of alternatives to it. The most famous case was Cybersyn in Chile, which was a project trying to put together socialism and cybernetics, which was interrupted by the coup d’etat funded by the US. There were a lot of projects similar to Cybersyn to create tech sovereignty for the region. But most of these projects were invisibilized or failed because of interruption or political pressures from other parts of the world. This politics of memory means trying to recover the invisibilized past of our regions related to community tech or alternative tech in order to imagine and codesign other possible digital futures.

Watch the full conversation between Rafael Grohmann and Alix Dunn here.