Action research involving six civil society organisations, seven provincial departments and 11 municipalities in Gipuzkoa shows that the ability to anticipate AI risks is unevenly distributed across local governance. The study proposes a place-based approach that combines digital inclusion, public participation and institutional coordination before AI systems become embedded in public services.
Artificial intelligence is entering public services, but local institutions do not all have the same ability to identify its risks, hear from affected communities or turn legal principles into everyday practice. A study of the Provincial Council of Gipuzkoa in the Basque Country argues that governing AI responsibly must begin with territorial digital inclusion: who can access digital services, who is represented in public data, and who has a voice in decisions.
Published in Law, Ethics & Technology , the study by Igor Calzada and Itziar Eizaguirre reports on an action-research programme carried out from 2024 to 2026. The programme connected six civil society organisations working with vulnerable communities, seven provincial departments and directorates, and 11 municipalities. It combined knowledge exchange, stakeholder engagement and open-science dissemination in an eight-step Science-for-Policy roadmap.
The central finding is that the capacity to anticipate the consequences of AI is unevenly distributed. Civil society organisations tend to recognise risks through lived experience of exclusion and discrimination. Provincial directorates are more likely to translate those concerns into formal tools such as audits, oversight frameworks and data-governance models. Municipalities, particularly those with fewer resources, face operational constraints, fragmented responsibilities and uncertainty about how to protect digital rights.
“The key question for public administrations is not simply whether to adopt AI, but whether they have the institutional capacity to govern it democratically,” says lead author Igor Calzada of the University of the Basque Country and Ikerbasque. “Our case study shows why foresight must be connected to digital inclusion, human rights and participation from the outset.”
Across the 11 municipalities, the researchers identified recurring concerns: dependence on mobile access, unclear responsibility for digital inclusion, limited visibility of vulnerable groups in official datasets, and uncertainty over responsibility for digital-rights protection. These differences mean that a single governance template cannot simply be copied from one territory to another.
The research instead presents anticipatory AI governance as a capability that administrations develop through repeated learning, coordination and engagement. Its proposed governance principles include foresight, transparency and explainability, participation and co-creation, human oversight, accountability, ethical and human-centred AI, and protection of digital rights.
“People who experience digital exclusion often see problems that administrative datasets cannot capture,” says co-author Itziar Eizaguirre of the Human Rights and Democratic Culture Directorate at the Provincial Council of Gipuzkoa. “Bringing civil society, municipalities and provincial departments into the same process helps make those experiences visible and relevant to policy design.”
The authors stress that Gipuzkoa is not a universal model. Its fiscal autonomy and administrative capacity create conditions that may not exist elsewhere. The transferable lesson is therefore not a fixed organisational design, but a set of principles—co-production, cross-departmental coordination and a rights-based approach—that must be adapted to local resources and institutions.
The study focuses on governance processes rather than the technical performance of particular AI systems, and it does not yet measure long-term policy outcomes. The authors call for comparative and longitudinal research to examine how anticipatory capacity develops in other regions, including places with fewer resources.
The paper, “Anticipatory AI governance for territorial digital inclusion strategies: action research within the Provincial Council of Gipuzkoa (Basque Country),” was published in Law, Ethics & Technology.
Calzada I, Eizaguirre I. Anticipatory AI governance for territorial digital inclusion strategies: action research within the Provincial Council of Gipuzkoa (Basque Country). Law Ethics Technol. 2026(3):0008, https://doi.org/10.55092/let20260008.
Law Ethics & Technology
Case study
Not applicable
Anticipatory AI governance for territorial digital inclusion strategies: action research within the Provincial Council of Gipuzkoa (Basque Country)
13-Jul-2026