What is Data Solidarity?

The collection, usage, sharing, and governance of medical data is plagued by a persistent problem of trust and equity. Too often, the communities providing data see little benefit, while corporations reap substantial rewards, further concentrating power and deepening public mistrust. This gap between risk, reward, and responsibility calls for new approaches to data governance. Data solidarity answers this call, aiming to transform how we govern health data, ensuring that trust, equity, and collective benefit are central, not peripheral, to data practices. In this blog, I reflect on a recent webinar hosted by the EthicsLab, where Professor Barbara Prainsack’s keynote on data solidarity was followed by a response from Dr. Nchangwi Munung.

The relationality of data

At its core, data solidarity challenges the unequal distribution of risks and benefits in the digital economy. It does this, amongst other things, by emphasising collective control and oversight, in addition to individual-level control over data. While individual consent remains important, it cannot address the structural problems of today’s data economies, such as structural inequalities. The proposed shift in regulation is connected to ontological commitments. As Prainsack explains, in today’s digitally networked reality, data is structurally relational because its meaning and value emerge only in networked, collective form. It is not simply about individuals, and often becomes meaningful and valuable only when aggregated, networked, and analysed in relation to other data. A single health record or genetic sequence tells us very little on its own; it’s only through comparison across populations, and the detection of patterns and correlations, that data gains predictive and commercial power. This means that even when data is collected from individuals, its consequences are collective—affecting whole communities and social groups, shaping public policy, guiding research priorities, and driving profit.

Solidarity-based data governance

Recognising this, data solidarity calls for a new foundation for data governance—one based on shared responsibility, collective oversight, and public accountability. It is structured around three key pillars that reframe data not as a commodity that individuals must constantly defend, but as a shared social resource that demands governance rooted in justice, equity, and interdependence.

The first pillar of data solidarity is about facilitating data use that creates significant public value without posing unacceptable risks. This means making it easier—legally, technically, and financially—for researchers, public institutions, and small entities to use data in ways that benefit society. It shifts the regulatory gaze from types of data to types of data use and evaluates each along two axes: public benefit and risk of harm. If a data use is high benefit and low risk, it should be actively supported through funding, streamlined access, or regulatory easing. If it’s low benefit and low risk, it may be allowed, but commercial actors should be required to share profits with the public. If it’s high benefit and high risk, it should be carefully assessed and only permitted if risks can be meaningfully reduced. And if it’s low benefit and high risk, it should be prohibited outright, with strong enforcement to deter harmful practices. To guide such decisions, researchers at the University of Vienna have developed PLUTO, a public value assessment tool, providing a structured framework for evaluating whether specific data uses create significant public value without posing unacceptable risks.

The second pillar addresses harm prevention and harm mitigation. Well-intentioned data use can still cause harm, be it through exclusion, discrimination, or other unanticipated consequences. Importantly, Prainsack emphasises that many harms in the digital age don’t stem from broken laws but from structural imbalances, relational inequalities, and opaque systems. This informs the proposal for harm mitigation bodies, conceptualised as accessible, responsive institutions that can offer redress and support, without requiring individuals to prove legal violations. Here, support is not contingent on litigation or blame but grounded in solidarity with those affected.

The third pillar calls for sharing the profits of corporate data use with the public. When companies build wealth by extracting and analysing data—data generated not just by individuals but by public infrastructures and collective social life—some of that value should return to the communities that make it possible. This could take the form of data taxes, benefit-sharing agreements, or reinvestment in public services. Crucially, this pillar distinguishes solidarity-based sharing from charity or Corporate Social Responsibility (CSR). Data solidarity is not a voluntary gesture of goodwill, but a structural obligation grounded in reciprocity and justice.

Finally, Prainsack draws a line between data solidarity and data sharing, arguing that whether data should be shared and used depends on its public value. While some data sharing reflects solidarity, not all sharing aligns with its goals. This distinction serves as a conceptual guardrail against moral blackmail, where the rhetoric of solidarity is misappropriated to guilt or coerce people into data sharing. Furthermore, the purpose matters more than the act. Data solidarity is focused on what the data is used for, not merely that it's being shared. The point then is to complicate a blanket and uncritical valorisation of openness that tends to characterise data sharing. This leaves open the possibility that solidarity may even require refusal to share, such as withholding data from actors who exploit or harm marginalised groups. In the final analysis, data solidarity is precisely the regulation required to prevent unencumbered data sharing from entrenching power asymmetries in health contexts.

In response to Prainsack’s presentation on data solidarity, Nchangwi Munung reflects on how solidarity might be operationalized in African contexts under conditions of regulatory constraint and uneven power dynamics. Munung’s response reframes two key claims made by Prainsack: the separation of data solidarity from data sharing, and the dismissal of corporate social responsibility (CSR) as a viable tool for benefit-sharing. In my view, the thrust of Munung’s response is the invitation to consider what a situated solidarity could look like. Dancing a tightrope between critique and pragmatism, her response begins from an attunement to the limits of what African institutions can currently demand from powerful commercial actors, articulating solidarity as something that must be continually worked out. This contextualisation of data solidarity can be read as a two-part intervention. If data sharing by corporate actors can be reframed as solidarity in data-scarce contexts (Move 1), then CSR might be a politically viable way to encourage such solidaristic sharing (Move 2).

The first move is to reclaim data sharing as solidarity in contexts of scarcity. Where data scarcity is a barrier to health equity, data sharing by commercial actors may function as a solidaristic act. Citing recent developments like Investec and Mediclinic offering genetic testing services to South African clients, she pointed out that these private companies are accumulating valuable genetic data in a region where data scarcity continues to be a major barrier to scientific progress and equitable genomic medicine. Here, the act of responsibly sharing data can play a corrective role.

Munung then shifts to the question of how solidaristic data sharing might be enabled in contexts lacking strong regulatory frameworks and considers Corporate Social Responsibility (CSR) — not as an ideal, but as a pragmatic mechanism for benefit sharing, especially where legal tools like taxation or enforceable governance are difficult to apply. Asking what is possible now, Munung posits that CSR could offer a way to encourage companies collecting valuable data to support local research, infrastructure, or data-sharing initiatives that benefit the public. To be clear, Munung doesn't frame CSR as a substitute for more robust forms of solidarity, but rather as a realistic point of entry.

This marked a productive tension in their dialogue, not a contradiction. Prainsack sets normative guardrails while Munung contextualizes data practices within African realities.

At the same time, this (CRS) self-policed, PR-driven, and non-binding corporate practice could, if leaned on too heavily, engender solidarity-washing, deflecting from the deeper demands of justice and accountability, thereby masking rather than mitigating extractivism. Furthermore, CSR approaches notoriously lack mechanisms for community-led governance. If companies define what counts as a ‘benefit’ or which communities to support, CSR can reproduce patronage logics and sideline the most affected groups.

Whether CSR might be a stepping stone toward more robust solidarity-based data governance is a pragmatic question worth asking, not an answer to settle on. But for CSR to be taken seriously within a solidarity framework, it would have to be deeply reimagined, tethered to community-defined priorities, and structured to avoid exploitation and capitulation to the profit imperative. Munung’s provocation, therefore, invites several questions. What would it mean to build a form of CSR that is grounded in reciprocity, fairness, and mutual obligation and not just in philanthropic branding? Can CSR evolve into a transitional instrument that supports the creation of public value until more robust governance architectures are in place?

Conclusion

Developed in response to the power asymmetries of the digital era, data solidarity challenges extractive data practices and advocates for mechanisms that ensure data serves the public good. In this vibrant discussion, Prainsack and Munung converge on several key principles: solidarity as a collective ethical orientation, public value and benefit-sharing as central goals, a broadened understanding of harm, the need for fair institutional mechanisms, and a shared critique of extractive, profit-driven data regimes. Their differences lie less in values than in emphasis and strategy, shaped by their distinct policy and research contexts — European and African respectively — making their exchange not a disagreement, but a rich and generative dialogue on how data solidarity might be realised in practice. This conversation on data solidarity forms part of a broader inquiry into ethics as worldmaking.