Data has become a foundational resource for contemporary societies. From scientific discovery and public health to climate action, economic development, and artificial intelligence, progress increasingly depends on the ability to access, share, and reuse data responsibly.

Yet while the importance of data continues to grow, the skills, institutions, and professional roles needed to steward data effectively are still emerging. Organizations across sectors are increasingly recognizing the need for dedicated expertise to manage data access, governance, and reuse in ways that generate public value while maintaining trust.

To help meet this need, we are pleased to launch DataStewards.net — dedicated to training and supporting data stewards.

Datastewards.net serves as the central hub for the training and capacity-building initiatives on data stewardship developed by The GovLab and The Data Tank. The site brings together programs, resources, and expertise designed to equip professionals with the practical knowledge required to enable responsible data sharing and collaboration.

Professional Courses and Training Programs
Datastewards.net offers targeted training opportunities for policymakers, researchers, civil society organizations, and companies seeking to strengthen their data stewardship capabilities, including:

• Data Stewardship Foundations: Bootcamps and hands-on programs co-developed with governments, foundations, and institutions to help participants design and implement real-world data collaboration initiatives.
• Deep Dives (Coming Soon): Short, focused masterclasses exploring specific aspects of data stewardship practice.
• Customized Solutions: Tailored offerings to institutionalize data stewardship across your organization.
• Exchange: A growing community of experts and alumni contributing to the development and delivery of training programs.

Together, these offerings aim to support the development of a global community of professionals capable of stewarding data responsibly and effectively — ensuring that data can be re-used to advance research, policy, and innovation while safeguarding rights and maintaining public trust.

Organizations and individuals interested in participating in training programs or collaborating on capacity-building initiatives can connect through datastewards.net.

Some Background

The launch of DataStewards.net builds on nearly a decade of work by The GovLab to advance the practice of data stewardship. Eight years ago, The GovLab launched the Data Stewards Network to promote responsible data leadership and support the reuse of data for the public good. In 2020, the Data Stewards Academy was established to provide dedicated training on responsible data access and reuse, and to broaden access to data stewardship training, Stefaan Verhulst collaborated with LinkedIn Learning in 2024 to develop an online course on data stewardship,

Since then, together with the Data Tank we expanded the initiative significantly:

• 8,500+ learners trained
• 103 guest faculty members contributing expertise
• Bootcamps and events held online and globally, including in Berlin, Brussels, Zürich, Milan and Annapolis

Datastewards.net marks the next phase of this effort, providing a dedicated platform to scale training and strengthen the global field of data stewardship.

RETIRED: Announcing the New Data Stewards Network

Eight years ago, The GovLab launched its Data Stewards Network. Developed with an understanding of the urgent need to open data for the public good, the Network served as both a means to advance responsible data leadership and a one-stop shop for the countless tools, events, and resources needed to build a community of practice that could use data to address the challenges of the 21st century. In 2020, we launched our Data Stewards Academy to offer dedicated training on data re-use.

Since then, the Network has grown to a scale we never anticipated. We have taught over 8,500 learners and recruited 50 guest faculty. We’ve hosted events across the world and run dedicated bootcamps in Berlin, Brussels, Zurich, and Annapolis.

To meet the growing demands of this program, we are excited to launch a new phase of this work and a new site to accompany it.

Screenshot of the new datastewards.net

DataStewards.Net is now the host of all of our data stewardship offerings at The GovLab and The Data Tank. This includes not merely our latest research but also a collection of our course offerings, a list of our faculty, and dedicated community resources. In the future, it will host dedicated Deep Dives — one-hour master classes for experienced data stewards hoping to learn more about a specific topic. We will launch this in the coming weeks, though interested learners can join our waiting list here.

We also offer information on our partners — which include Microsoft, the Bertelsmann Foundation, and State of Maryland, Cariplo Foundation, Fondazione Compagnia di San Paolo, Hasler Stiftung, and more. Our partner’s support has been incremental in helping us set up and scale our data stewardship efforts. Both individuals and organizations interested in collaborating with the Data Stewards Network, either by participating in a course or seeking their own dedicated training, can do so through a specialized contact page.

While we will continue to maintain our Medium and other publications for the short term, we plan to make this site our central platform going forward. We hope you will visit it regularly and subscribe to our newsletters to stay up to date.

We invite you to check out all on offer at the improved datastewards.net. If there’s anything we can do to improve our offerings, reach out to contact@datastewards.net and let us know.

Introducing DataStewards.net: Advancing the Practice of Data Stewardship for an AI Era was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

Data governance has become a catch-all term, used to describe everything from data quality and metadata to privacy, compliance, and digital strategy. This ambiguity is not harmless: when concepts remain unclear, responsibilities blur and decisions stall. Demystifying data governance therefore requires two things: clarifying how it relates to neighboring concepts, and understanding the components and trends that are reshaping governance in an AI- and data-intensive world.

Our new resource, What Is Data Governance? 30 Questions and Answers, is designed to demystify the concept. Instead of offering yet another definition, it takes one already tested through the Broadband Commission’s Data Governance Toolkit and it clarifies how data governance relates to terms like data management, stewardship, privacy, and compliance.

The Q&A uses the same working definition as the Toolkit: data governance involves the processes, people, policies, practices and technology that seek to govern the data lifecycle toward meeting the purpose of increasing trust, value, and equity, while minimizing risk and harm in alignment with a set of core principles.

While this may sound abstract, the entire Toolkit makes it actionable.

The definition breaks governance into four questions any organization must answer:

• Why — What vision and purposes should data governance serve here?
• How — Which principles will guide decisions and implementation of practices across the lifecycle?
• Who — What people carry which responsibilities, and with what legitimacy?
• What — What concrete policies, procedures, and technologies will we implement to meet the purpose and comply with the principles?

The Q&A runs through six sections. Foundations clarifies the differences between governance and adjacent concepts such as strategy, management, ethics, privacy, and AI governance — often conflated in practice. Later sections cover purpose and vision, principles, legitimacy and agency (including Indigenous data sovereignty and Digital Self-Determination), participation and accountability, lifecycle practices, sector-specific regimes, cross-border flows, and emerging trends. Throughout, the emphasis is on operationalization: translating principles into processes, roles, contracts, and tools.

Why now?

Because organizations are under simultaneous pressure to use data more intensively (innovation, AI, evidence-based policy, crisis response) while navigating expanding legal, ethical, and geopolitical constraints. Governance is no longer an IT niche: it’s central to public trust, regulatory scrutiny, and the capacity to reuse data responsibly across silos and borders.

This Q&A fills that gap:

• As a primer for those needing an integrated overview
• As a workbench for practitioners designing frameworks or running workshops
• As a translation layer connecting debates on Indigenous data sovereignty or Digital Self-Determination with mainstream governance discourse

By aligning with the Data Governance Toolkit and international guidance, it sits within a broader shift toward human-rights-based, stewardship-oriented governance that respects subsidiarity, proportionality, and sectoral realities.

A living document: your feedback

This Q&A isn’t meant as the final word. The 30 questions emerged from real engagements with policymakers, regulators, practitioners and researchers and they’re almost certainly incomplete. New technologies, business models, and legal developments will raise fresh questions.

We see this as a living document that should evolve alongside practice. We especially welcome feedback on:

• Clarity: Where are answers too abstract? Which distinctions remain blurry?
• Gaps: Which questions are missing, particularly from under-represented regions, communities, or institution types (small municipalities, community initiatives, Global South organizations)?
• Examples: Which case studies, playbooks, or tools have helped you move from principles to implementation?

Your insights can inform future iterations, including refining entries, adding new questions, and expanding the tools annex. This mirrors the governance model the GovLab advocates: treating norms as evolving arrangements that are continuously tested in practice, contested when necessary, and adapted in light of new evidence and experience.

If you have thoughts, critiques, or examples, please share them. Data governance will never fit into a single, perfect definition, but together we can build a shared, practical language for governing data in ways worthy of the trust we ask for.

Click here to access the resource.

Data Governance Without the Jargon: 30 Questions and Answers to Clarify Terms and Trends was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

As part of an ongoing effort to contribute to current topics in data, technology, and governance, The GovLab’s Selected Readings series provides an annotated and curated collection of recommended readings on themes such as open data, data collaboration, and civic technology.

In this edition, we curate writings from the last two years on Indigenous data sovereignty and Indigenous data governance. It complements a previous iteration on these topics.

Key takeaways:

• Data sovereignty as self-determination: Indigenous communities seek authority over how data about them is collected, used, and shared.
• Broader definitions of data: Data includes culture, language, land, and relationships — not just statistics.
• Relational governance: Trust, co-creation, and long-term partnerships are essential, not one-off consent.
• CARE and Indigenous frameworks matter: Governance models centered on collective benefit, control, and ethics are reshaping practice.
• AI is a double-edged sword: It creates risks of extraction and bias but also opportunities, especially for language and cultural preservation.
• Implementation gaps remain: Legal recognition, capacity, standards, and institutional models are still under development in many contexts.

To suggest additional readings on this or any other topic, please email info@thelivinglib.org. All our Selected Readings can be found here. All readings listed in alphabetical order.

***

Selected Reading List (in alphabetical order)

• Australian Government. 2024. “Framework for Governance of Indigenous Data (GID) | NIAA.” National Indigenous Australian Agency, May 30.
• Caballar, Rina Diane, and Sophie Bushwick. 2023. “How Indigenous Groups Are Leading the Way on Data Privacy.” Scientific American, June 7.
• Calac, Alec J., and Timothy K. Mackey. 2025. “A Systematic Review of Responsible Stewardship of Research and Health Data from Indigenous Communities.” Npj Digital Medicine 8 (1): 504.
• Cunningham-Reimann, Stephanie, Aarti Doshi, Sterling Stutz, Gabriel B. Tjong, and Angela Mashford-Pringle. 2025. “Flipping Data on Its Head: Differing Conceptualisations of Data and the Implications for Actioning Indigenous Data Sovereignty Principles.” Global Public Health 20 (1): 2450395.
• Danton, Cheryl M., and Christopher Graziul. 2026. Confronting the Challenges of Sensitive Open Data. January 6.
• Department of Health and Disability and Ageing (Australia). 2026. Governance of Indigenous Data: Implementation Plan. January 23. Australia.
• Drake, Allison K., Paulatuk Hunters and Trappers Committee, Tony Green, et al. 2025. “Advancing Indigenous Data Governance through a Shared Understanding in Paulatuk, Inuvialuit Settlement Region.” Frontiers in Environmental Science 13 (March).
• Hurst, Samantha, Kelle Dhein, Joseph Yracheta, and Timothy Mackey. 2026. “Sovereignty in the Digital Age Indigenous Perspectives on Health Data and Emerging Technologies.” SSM — Qualitative Research in Health 9 (June): 100698.
• Jennings, Lydia, Katherine Jones, Riley Taitingfong, et al. 2025. “Governance of Indigenous Data in Open Earth Systems Science.” Nature Communications 16 (1): 572.
• Khani, Mohammad Milad. 2025. “Data Repurposing, Algorithmic Bias and Indigenous Privacy in the Age of AI | IAPP.” IAPP.Org, IAPP, November 5.
• Ray, Shawn, and Ram L. Ray. 2026. “Incorporating Indigenous Knowledge Systems into AI Governance: Enhancing Ethical Frameworks with Maori and Navajo Perspectives.” AI and Ethics 6 (1): 114.
• Suchikova, Yana, and Serhii Nazarovets. 2025. “Extending the CARE Principles: Managing Data for Vulnerable Communities in Wartime and Humanitarian Crises.” Scientific Data 12 (1): 420.
• Tanner, Brooke, and Cameron Kerry. 2025. “Can Small Language Models Revitalize Indigenous Languages?” Brookings, March 19.
• Te Kāhui Raraunga (2025). Māori Artificial Intelligence Governance Framework. Contextualised advice for AI use, extending the Māori data governance model; Kukutai, T., Campbell-Kamariera, K., Mead, A., Mikaere, K., Moses, C., Whitehead, J. & Cormack, D. (2023). Māori data governance model. Te Kāhui Raraunga.
• “Consultation on Indigenous Data Governance Workshop, on Development Dimensions of Indigenous Traditional Knowledge, Cultural Heritage, and Biodiversity Data.” 2025. UN Data Governance Working Group, October 21.
• “How Indigenous Knowledge Drives Scientific Discovery.” 2026. The UNESCO Courier, January.

Annotated Selected Reading List (in alphabetical order):

Australian Government. 2024. “Framework for Governance of Indigenous Data (GID) | NIAA.” National Indigenous Australian Agency, May 30.

• This is a framework for working with Indigenous data that seeks to place Aboriginal and Torres Strait Islander people at its core and was co-designed with Aboriginal and Torres Strait Islander partners.
• It outlines a seven-year implementation plan for supporting increased agency of First Nations Australians line line with four principles: partner with Aboriginal and Torres Strait Islander people; build data-related capabilities; provide knowledge of data assets; build an inclusive data system.
• The framework “acknowledges that better outcomes are achieved if First Nations people have a genuine say in matters affecting them — including the use of data to inform policy-making in government” and calls for data-related capacity building in these communities

Caballar, Rina Diane, and Sophie Bushwick. 2023. “How Indigenous Groups Are Leading the Way on Data Privacy.” Scientific American, June 7.

• This article argues that Indigenous communities are leading the way in development of privacy-first data storage systems that give users control and agency over their information. Organizations like Terrastories, Our Data Indigenous, and Āhau do not rely on the cloud but on other methods because it “is almost like the knowledge is leaving the territory because it’s not physically present”.
• The article also notes that some Indigenous-led apps offer more customized and granular levels of access and permissions. Terastories, for example, only lets maps and stories be viewable by members who have logged in to the app using their community’s credentials. Adding or editing requires editor access.
• These are further supplemented by codes of conduct that researchers must follow to get access to data. Āhau, for example, adheres to the Te Mana Raraunga principles of Māori data sovereignty that requires giving Māori communities authority over their information and acknowledging the relationships they have with it.

Calac, Alec J., and Timothy K. Mackey. 2025. “A Systematic Review of Responsible Stewardship of Research and Health Data from Indigenous Communities.” Npj Digital Medicine 8 (1): 504.

• This is a systemic review that aims to detail Indigenous Data Sovereignty considerations and practices in health research in the United States and elsewhere and to identify frameworks that operationalize IDS practices for responsible research and use of data-driven technologies. The review finds 41 relevant articles detailing specific considerations around collection, access, and use of Indigenous data as well as various regional or cultural considerations.
• The authors identify specific data management considerations related to data access, ownership, knowledge systems and relationality; responsible data governance, metadata, and regulatory authority; and open-access challenges and tribal leader engagement.
• It also notes how new technologies and systems are being rapidly designed, often without explicit rules and regulations to operationalize indigenous data sovereignty and protect Indigenous Peoples. It notes that digital technologies can benefit communities but only when they are used in a way that respects the communities that provide foundational data.

Cunningham-Reimann, Stephanie, Aarti Doshi, Sterling Stutz, Gabriel B. Tjong, and Angela Mashford-Pringle. 2025. “Flipping Data on Its Head: Differing Conceptualisations of Data and the Implications for Actioning Indigenous Data Sovereignty Principles.” Global Public Health 20 (1): 2450395.

• This article offers commentary on how to integrate Indigenous data sovereignty principles into data work. The authors begin by definition data in Western research (as “tangible, measurable, and observable phenomenon devoid of observer bias and contextualisation”) and Indigenous contexts (as information “created by, for and with Indigenous Peoples, communities, and Nations and includes language, culture, ceremonies, lands, dreams, songs, drums, instruments, artwork, clothing, regalia, beadwork, as well as stories, interviews, transcripts,artifacts (historic and contemporary), and statistic”).
• The authors then outline how Indigenous data sovereignty is “predicated on distinct worldviews, cultures, histories, and data stewardship” and center an the overarching principle of self-determination in “data collection, analysis, usage,knowledge translation and dissemination, with Indigenous Peoples’ interests and decision-making authority in determining who can access and analyse Indigenous data and for what purposes”. Researchers engaging in Indigenous research must collaborate equitably with Indigenous communities throughout the research process, starting with research design and ending with dissemination.
• The authors then offer several recommendations to researchers, scholars, policy makers, government staff, and other allies interested in implementing Indigenous data sovereignty. These include completing training from the First Nations Information Governance Centre; undertaking cultural safety training that includes socio-political history and developing authentic relationships with indigenous peoples; questioning the research process by co-creating the research process with Indigenous peoples; building relationships that lead to co-creation and are reciprocal beyond the initial project; and to apply a distinctions-based lens that respects the unique needs and characteristics of the specific Indigenous community they are working with.

Danton, Cheryl M., and Christopher Graziul. 2026. Confronting the Challenges of Sensitive Open Data. January 6.

• This article introduces the term “sensitive open data” to describe “data containing private information about individuals that becomes available to the public through a range of mechanisms, from court orders to statutory mandates.” It highlights lessons that might be gleaned from Indigenous communities.
• It describes the FAIR Principles and the complementary CARE Principles as guiding resources to “integrate Indigenous worldviews that center ‘people’ and ‘purpose’ to address critical gaps in conventional data frameworks by ensuring that Indigenous Peoples benefit from data activities and maintain control over their data.
• The authors highlight two use cases that demonstrate the value of the CARE Principles, a 2014 project by the National Congress of American Indians Policy Research Center to improve census participation and the first tribal IRB, the Navajo Nation Human Research Review Board, which exercises sovereignty over all human research activities in the Navajo Nation.

Department of Health and Disability and Ageing (Australia). 2026. Governance of Indigenous Data: Implementation Plan. January 23. Australia.

• This document is an implementation plan put forth by the Australian Government’s Department of Health, Disability, and Aging to require all entities to embed the Australian Government Framework for governance of Indigenous data, which guides users to partner with Aboriginal and Torres Strait Islander people (APS), build data-related APS capabilities, provide knowledge of data assets, and build an inclusive data system.
• Acknowledging that the Department holds substantial datasets about the Australian Government, including APS, the department commits to working on genuine partnership “to ensure that First Nations people are afforded the right to exercise ownership and control over Indigenous data across all phases of the data lifecycle.”
• The plan then outlines various goals and actions to achieve this aim, including efforts to “embed governance of Indigenous data in department policies and processes”, “build and maintain meaningful partnerships with First Nations people, communities, and organizations”, “develop and implement methods for First Nation people to know what data are held relating to their interests, its use, and how it can be accessed”, “build governance of Indigenous data capability for department staff and First Nations partners”, “support and engage in organizational and cultural change to improve governance of Indigenous data”, and “monitor and evaluate governance of indigenous data implementation”.

Drake, Allison K., Paulatuk Hunters and Trappers Committee, Tony Green, et al. 2025. “Advancing Indigenous Data Governance through a Shared Understanding in Paulatuk, Inuvialuit Settlement Region.” Frontiers in Environmental Science 13 (March).

• This piece examines Indigenous data governance frameworks in the Canadian arctic. To support Indigenous self-determination, Fisheries and Oceans Canada researchers and the Paulatuk Hunters and Trappers Committee co-developed a data governance statement specific to an interview project. This report then details the steps and dialogue that characterized the creation of this statement and five considerations to strengthen future efforts.
• The piece then outlines a sequence that took place over several months where the research team developed academic ethics resources and documents, which were then reviewed by Indigenous group leaders. The authors provide a summary of the materials, which are living documents and can be updated as needed when applicable to other projects.
• The piece then suggests, based on the project experiences, that future researchers center community and project context, the changing digital landscape, individual and collective knowledge protections, planned project outputs, and confidentiality and anonymity nuances. It concludes with a note that indigenous data governance is “integral to mutually-beneficial research partnerships that are necessary in an era of rapid environmental change.”

Hurst, Samantha, Kelle Dhein, Joseph Yracheta, and Timothy Mackey. 2026. “Sovereignty in the Digital Age Indigenous Perspectives on Health Data and Emerging Technologies.” SSM — Qualitative Research in Health 9 (June): 100698.

• This study examines how Indigenous experts interpret the promises and risks of emerging digital technologies in relation to data sovereignty. It explores how Indigenous communities understand data governance challenges, evaluate technological innovations, and identify cultural, legal, ecological, and infrastructural considerations that shape sovereignty in digital contexts.
• The researchers identify four central themes for practitioners to be cognizant of to promote Indigenous data sovereignty: (1) The Significance of Indigenous Data Sovereignty; (2) Balancing Sovereignty, Trust, and Capacity in Health Data Management; (3) Integrating Native Values into Data Governance and Technology Systems; and (4) Technology and Tribal Sovereignty: Opportunities and Challenges.
• The piece concludes with a note that this work affirms and extends current Indigenous data sovereignty frameworks, “while making visible the deeper infrastructural and jurisdictional challenges that shape technology use in Indian Country. Addressing these challenges is essential for building the data foundations that allow Tribal Nations to design responsive health interventions, monitor outcomes, and strengthen the cultural relevance of health system decision-making.”

Jennings, Lydia, Katherine Jones, Riley Taitingfong, et al. 2025. “Governance of Indigenous Data in Open Earth Systems Science.” Nature Communications 16 (1): 572.

• The Earth Data Relations Working Group convened to discuss what processes are needed to follow open science protocols while upholding Indigenous Peoples’ rights. The piece begins by reviewing open science mandates but notes, “mandates and their implementation don’t often consider the rights of Indigenous Peoples in relation to Earth observation data” and there are few examples of what application of Indigenous knowledge systems looks like in an open science and repositories context.
• It notes that many Indigenous worldviews center relationships as a core value and that “relationships exist at the scale of the individual and collectives and apply to connections between humans, places, and biotic and abiotic communities in the past, present, and future.”
• It recommends Earth science researchers adopt four practices to implement the CARE Principles and advance Indigenous Data Governance. First, the piece suggests adopting ongoing, internal learning and assessment practices to ensure that CARE is adopted as an iterative process and not treated as a one-off activity. Second, it suggests that organizations develop collaborative access permission and research protocols with Indigenous Peoples; enhance data and specimen management infrastructure to manage samples and specimens in ways consistent with Indigenous communities’ needs and expectations (including applying metadata and context labels); and enhancing data relations at all stages of the data lifecycle, which includes engaging data hosts on Indigenous metadata and amending citations.

Khani, Mohammad Milad. 2025. “Data Repurposing, Algorithmic Bias and Indigenous Privacy in the Age of AI | IAPP.” IAPP.Org, IAPP, November 5.

• This article looks at Indigenous data sovereignty in Canada, which has a treaty obligation to recognize First Nations data sovereignty. However, the federal government has yet to yet to explicitly and fully recognize First Nations data sovereignty in legislation. This is a problem amid growing demand for datasets to train AI.
• The author detail the ongoing threats to privacy and data sovereignty. There are growing concerns about the repurposing of personal information and what happens to Indigenous personal information that is reused for a secondary objective. It notes that ownership, control, access, and possession play an important role in data sovereignty.
• The author recommends that Canada’s Personal Information Protection and Electronic Documents Act should be amended to include clauses specifically about protecting the privacy of First Nations. The requirement for collective consent should be established alongside individual consent when data involves Indigenous communities. The author further argues that data sovereignty must be explicitly defined within the legal text, not relegated to secondary sources.

Ray, Shawn, and Ram L. Ray. 2026. “Incorporating Indigenous Knowledge Systems into AI Governance: Enhancing Ethical Frameworks with Maori and Navajo Perspectives.” AI and Ethics 6 (1): 114.

• This paper examines the limitations of Western-centric ethical frameworks through a comparative analysis of Māori Kaitiakitanga and Navajo Hózhó principles, offering a framework for integrating these principles into AI governance.
• Māori Kaitiakitanga emphasizes guardianship, responsibility, and the interconnectedness of all living things. By recognizing the right of all natural entities, Kaitiakitanga can inform the development of AI systems that prioritize the long-term health of ecosystems and challenges the assumption that all technological expansion is inherently beneficial.
• Navajo Hozho, a core principle in Navajo philosophy, emphasizes “harmony, balance, beauty, and the interconnectedness of all beings. It represents a holistic worldview that seeks to maintain equilibrium between physical, mental, spiritual, and social well-being”. It promotes AI that is ethically sound, culturally sensitive, and aligned with the values of community and environmental well-being. The recognition of interconnectedness is crucial for decisions regarding data collection, algorithm design, and AI deployment.
• The authors argue that while there is some tensions between the concepts — for example the emphasis on collective ownership in Kaitiakitanga — they offer a way forward in addressing AI governance challenges. Navigating these tensions requires open dialogue between Indigenous communities, AI developers, and policymakers and a critical examination of Western ethical principles and legal frameworks.

Suchikova, Yana, and Serhii Nazarovets. 2025. “Extending the CARE Principles: Managing Data for Vulnerable Communities in Wartime and Humanitarian Crises.” Scientific Data 12 (1): 420.

• This essay examines the CARE Principles (Collective Benefit, Authority to Control, Responsibility, Ethics) that were developed for the ethical stewardship of Indigenous data and to complement the technical standards of FAIR (Findable, Accessible, Interoperable, Reusable). The authors argue that the CARE Principles are a useful model for data governance for other vulnerable communities — such as those in conflict-affected environments.
• Looking through the lens of the Russo-Ukrainian war, the piece notes how CARE can be adapted to new contexts. Collective benefit can be used to ensure data on IDPs and residents of occupied territories is used for humanitarian assistance, family reunifications, and long-term recovery. Authority to control can be used to ensure that those in conflict maintain autonomy over their personal information. Responsibility can demand that those using data in the Russo-Ukrainian context minimize harm, ensure informed consent, and foster trust with affected communities. Ethics, finally, can call organizations to commit to prioritizing the immediate and long-term recovery of at-risk groups.
• The authors conclude by stating that extending CARE in this way does not compromise their original intent, to empower Indigenous communities, but underscores that universality and relevance as a robust ethical framework. They argue, “the global significance of CARE lies in its capacity to establish a standardized ethical framework for data governance across a wide range of contexts.”

Tanner, Brooke, and Cameron Kerry. 2025. “Can Small Language Models Revitalize Indigenous Languages?” Brookings, March 19.

• This article examines how small language models may offer a cost-effective and resource-efficient solution for indigenous communities seeking to preserve their language, which is critical to preserving a cultural identity and transmitting unique worldviews, traditions, and knowledge.
• The authors argue that “SLMs can respond to, analyze, and translate a variety of languages that […] are not widely used or recorded in digital formats.” They do this by reducing computational and data requirements in areas where high-end infrastructure is not available. It then highlights several examples of how SLMs have been successfully adapted specifically for Indigenous languages — designed, trained, and fine-tuned with input from the communities they serve.
• The piece then offers guidance on collaborating on Indigenous communities, noting that community input and participation through the data collection, model training, and subsequent deployment can ensure that models respect traditional knowledge and align technological outcomes with community values. It specifically highlights one example of data stewardship from Te Hiku Media, a small nonprofit radio station in New Zealand, which created data licensing agreements to ensure that projects created with Māori data will benefit the Māori people as a whole.

Te Kāhui Raraunga (2025). Māori Artificial Intelligence Governance Framework. Contextualised advice for AI use, extending the Māori data governance model; Kukutai, T., Campbell-Kamariera, K., Mead, A., Mikaere, K., Moses, C., Whitehead, J. & Cormack, D. (2023). Māori data governance model. Te Kāhui Raraunga.

• This is a comprehensive framework designed by Māori AI experts to guide the use of artificial intelligence across Aotearoa New Zealand public service. It recognizes that AI is “data-driven intelligence” and therefore needs to be culturally grounded in ways that “protect Māori data sovereignty and uphold te Tiriti o Waitangi. Without culturally grounded and Tiriti-anchored governance, there is a risk of systemic harm and extraction rather than partnership and benefit.”
• It centers several key values including “nurturing data as a taonga that requires culturally grounded models of protection and care”, putting iwi-Māori data in iwi-Māori hands, using data for good, being accountable, and decolonizing data ecosystems.
• It highlights several desirable outcomes including “reaffirming and strengthening connections to identity, place and te reo Māori” and “supporting whānau to flourish”.

“Consultation on Indigenous Data Governance Workshop, on Development Dimensions of Indigenous Traditional Knowledge, Cultural Heritage, and Biodiversity Data.” 2025. UN Data Governance Working Group, October 21.

• This report addresses the challenges of data governance in Sarawak, Malaysia. While major steps have been taken to foster digital inclusion, one crucial element remains to be developed: Indigenous data governance. CARE and FAIR provide important foundations, they need to be localized to meet the needs of the people of Sarawak.
• The ability to achieve Indigenous data governance is limited by a lack of contextual governance models that align with customary law, the diverse nature of Indigenous data, the ethical challenges that come from digitalization, interoperability and institutional silos, weak national policy integration, an absence of clear protocols for Indigenous data governance, and gaps in legal protection and benefit-sharing mechanisms.
• To address these and other gaps, the report offers several lessons from case studies around the world. It then offers a list of policy recommendations, such as institutionalizing the CARE Principles at the State Level, establishing a Sarawak Indigenous Data Governance Framework, develop Sarawak-specific metadata standards, pilot interoperable data spaces, embed tiered consent frameworks, secure UN technical support, develop a community consent and data governance toolkit, and promote ethical data literacy across stakeholders.

“How Indigenous Knowledge Drives Scientific Discovery.” 2026. The UNESCO Courier, January.

• This report from UNESCO highlights the ways in which indigenous knowledge can facilitate scientific discovery. It offers various articles offering lessons from Brazil on saving a fish species with the Paumari people, health research in China that relies on Dai medicine, astronomical research in Kalahari with Jul’hoansi hunter-gatherers, and efforts to adapt to climate with the Sami people.
• Each of these cases demonstrate how Local and Indigenous Knowledge Systems can be mutually beneficial. It demonstrates the need for exchanges between Indigenous knowledge holders and scientists to foster better empirical data.
• The authors emphasize that any exchange must be accompanied by guarantees of free and informed consent and equitable sharing of benefits. They argue that, “the objective is not simply to absorb Indigenous knowledge into mainstream science, but to promote cooperation for the benefit of all.”

Selected Readings on Indigenous Data Governance: 2026 Update was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

On January 8, 2026, The GovLab and the Open Data Policy Lab hosted “Looking Back to the Future” — a conversation that looked at the forces that shaped 2025 and may define 2026 across data, AI, and governance.

Moderated by Stefaan Verhulst (Co-Founder, The GovLab; Research Professor, NYU Tandon School of Engineering), the webinar brought together three panelists offering different perspectives on the evolving data and AI landscape:

• Jed Sundwall, Executive Director, Radiant Earth
• Guilherme Canela De Souza Godoi, Director, Division for Digital Inclusion and Policies & Digital Transformation and Secretariat of UNESCO’s Information for All Programme
• Susan Ariel Aaronson, Research Professor of International Affairs and Director of the Digital Trade and Data Governance Hub (and Principal/Co-Principal Investigator with NSF and NIST on responsible AI)

A clear framing emerged early: While AI is experiencing a kind of “summer,” the world may also be entering a “data winter” — marked by a rapid decline in access to data for public interest purposes. While new actors might fill gaps in data access, this required updating access regimes and rethinking basic institutional foundations underlying data systems.

The 2025 Rearview Mirror: What Stood Out

1) Access to high-impact data is increasingly shaped by non-state actors

The conversation began with Sundwall highlighting how sharply declining resource and technology costs are enabling smaller nonprofits and research institutions to undertake activities that were once the exclusive domain of states.

Sundwall pointed to the emergence of Common Space, a nonprofit established in 2025 to launch a humanitarian satellite and the Environmental Defense Fund’s MethaneSAT. Both serve as evidence that barriers to entry in space-based data collection have fallen dramatically. As he noted, these initiatives emerge in part from a recognition that “a lot of the actors who have access to high-resolution satellites and imagery and can control those things might not prioritize humanitarian efforts.”

This shift, he suggested, reflects less a deliberate move away from governments than a response to declining trust, access, and prioritization within some state-controlled data systems, with falling costs in computing, sensors, and satellite technology serving as the primary mitigating factor enabling non-state actors to step in — expanding the range of institutions capable of producing planetary-scale datasets, rather than eliminating the role of governments altogether.

2) AI can facilitate access to data — not just depend on it

Sundwall then emphasized that AI can play a role not only in consuming data, but also in making certain kinds of data easier to access and use. One example he highlighted was the use of AI-generated embeddings derived from satellite imagery — essentially compact representations of images rather than the full images themselves. As he explained, “embedding outputs trained on satellite imagery are just much, much smaller than all of that imagery that people normally have to sift through,” which means they “stand to make much more data much, much more accessible to people.”

Instead of requiring users to store, move, or analyze massive volumes of raw geospatial imagery, these embeddings allow people to work with far smaller data outputs that still retain useful information. Because they are significantly smaller, they can be easier to move, share, and use, potentially lowering technical and resource barriers to working with large-scale geospatial datasets. At the same time, Sundwall cautioned that important questions remain unresolved, including “how reliable are they?” and “what are they even telling us?” He noted that embeddings can be difficult to interpret and understand in practice.

He also pointed to a more operational role for AI: reducing the friction involved in adopting data standards. AI tools, particularly large language models that assist with writing and adapting code, can help automate what he described as “the boring chore work” of implementing standards, making it easier for more organizations to use shared standards that support interoperability and reuse.

3) Multilateral governance is undergoing “tectonic” change

Speaking from the perspective of multilateral institutions, Canela described 2025 as a year of “tectonic changes” in global governance. By this, he meant a break from long-standing assumptions that have shaped the multilateral system since World War II, particularly the expectation that states would broadly cooperate, share responsibility, and work through international institutions to address global challenges. Instead, he noted that some major state actors were increasingly stepping back from these roles or openly questioning the value of multilateral organizations themselves.

These shifts, Canela explained, made decision-making much harder. Established processes that once relied on consensus and predictable participation could no longer be taken for granted, and negotiations increasingly unfolded in a more fragmented and contested environment. As he put it, 2025 was “a very difficult moment for the decision-making process,” marked by what he described as “a clear attack against the idea of doing multilateral policy and multilateral governance.”

At the same time, Canela pointed to a more hopeful development, particularly within the UN system and regional organizations. Amid intense “AI hype” and a rush by individual agencies and institutions to launch their own initiatives, there was growing recognition that uncoordinated efforts risked duplication and confusion. Over the course of 2025, he observed signs of greater willingness among these actors to coordinate their approaches and align their work.

He highlighted the WSIS+20 review process as a concrete example of this alignment. One of the few UN General Assembly resolutions adopted by consensus, the review reaffirmed that technology governance should remain firmly grounded in human rights. Importantly, Canela emphasized that the central challenge is not how to regulate each new technology as it emerges, but how to ensure that the broader governance frameworks guiding digital transformation are rights-based, coherent, and fit for purpose. In his words, the focus should be on “the overall framework,” rather than on individual technologies in isolation.

4) Data sovereignty, national security, and cross-border flows are intensifying fault lines

Aaronson concluded the review of 2025 by highlighting two developments over the past year that illustrate diverging approaches to data governance and access.

First, she discussed China’s efforts to create formal data markets — systems intended to make buying and selling data more structured and transparent. While the idea has attracted significant policy attention, Aaronson was blunt about the results. As she put it, China’s data markets have been “shockingly unsuccessful so far” and have “not gotten a lot of companies to participate,” despite repeated attempts to make them work.

She also described China’s approach to AI governance as focusing on how AI systems are used and who uses them, rather than on the technology alone. This includes identifying specific chatbot uses considered harmful and issuing draft rules for public comment. Aaronson highlighted this as unusual in context, noting that “the Chinese government actually put the draft rules out for public comment,” which she described as “quite interesting” given the country’s broader political environment.

Turning to the United States, Aaronson emphasized a broader shift toward restricting access to data, with national security increasingly used as the justification. She pointed to a 2024 decision, made before President Trump returned to office, in which the U.S. imposed its first explicit limits on cross-border transfers of certain categories of sensitive personal data to six countries designated as “terrorist nations.” For Aaronson, this move was significant not only because of the terrorism framing, but because it marked “the first time the United States has really restricted data flows in this way.”

She warned that these measures are part of a wider trend toward reduced data access. When combined with expanding “data sovereignty” policies, she argued, they risk sharply limiting global data sharing, an outcome she described as deeply concerning, given that “science and innovation depend on data that crosses borders.”

Looking ahead: 2026 themes the panel expects to intensify

1) New data institutions beyond the state

A central expectation for 2026 is the growing role of non-state actors — such as research labs, nonprofits, and consortia — in producing datasets that increasingly function as de facto public infrastructure.

Jed Sundwall pointed to large-scale examples already emerging. One was a planetary-scale, 3D building footprint dataset produced by a research lab at the Technical University of Munich. Another was the Overture Maps Foundation, a consortium of major technology companies working to create a shared global mapping dataset and a system of unique identifiers for real-world entities.

Sundwall argued that these efforts signal a deeper shift: data critical to public decision-making is no longer produced only by governments, yet governance models have not kept pace. As he put it, “We’re going to need new kinds of institutions that can actually produce and govern data properly,” adding that these institutions are unlikely to be “purely governmental” or “purely private.”

To illustrate what he meant, Sundwall referenced existing global coordination bodies such as ICANN (which manages the internet’s domain name system) and GS1 (which governs global barcode standards). These organizations do not belong to any single government, yet they provide trusted infrastructure that underpins global commerce and connectivity. In his view, comparable institutional models may be necessary to govern shared data resources responsibly at planetary scale.

2) Rethinking “open”: from a slogan to design choices

Sundwall cautioned that “open” is often treated as a catch-all label rather than a precise concept. He described it as a sloppy or imprecise term, noting that openness is only one attribute of a data product, not a guarantee of quality, ethics, or sustainability. For clarity, he urged moving away from treating “open” as a goal in itself and instead asking more fundamental questions: Is the data useful? Is it produced and used ethically? Who pays to maintain it over time? Who captures the value it generates?

Guilherme Canela reinforced this point by calling “open” an “omnibus concept”. By this, he meant that without governance models, context, and purpose, the word risks obscuring real trade-offs rather than clarifying them. As data ecosystems expand beyond governments to include private firms, nonprofits, and hybrid actors, he argued, governance must become more explicit about how openness is designed and for whom it works.

Both speakers suggested that 2026 may mark a turning point: a recognition that producing high-quality, widely usable data is costly, and that older assumptions about unlimited openness may need revision, particularly when large, well-capitalized actors extract disproportionate value from shared resources.

3) Updating access-to-information regimes for a non-governmental data world

Looking ahead, Canela highlighted a symbolic milestone: 2026 marks 260 years since the world’s first Freedom of Information (FOI) law, adopted in Sweden in 1766. He used this anniversary to pose a forward-looking governance question: how should access-to-information frameworks evolve in a world where governments are no longer the primary holders of socially important data?

Canela urged for organizations to adapt long-standing principles — such as transparency and public access — to current realities. While roughly 140 FOI laws exist globally, he noted that only a small number meaningfully apply to private or non-state actors, with South Africa standing out as a notable example.

In 2026, he argued, the challenge will be whether access-to-information regimes can be redesigned to reflect today’s data ecosystem — one in which corporations, platforms, and research institutions play central roles — without abandoning the democratic principles that FOI laws were meant to protect.

4) “Quality inputs” and the risk of a data tragedy of the commons

Canela also warned that weakening trusted sources of information could have cascading consequences across society. If science, independent journalism, and reliable data producers are undermined, he argued, the entire information ecosystem suffers — affecting not only governance but also business and innovation.

He framed this risk as a potential tragedy of the commons: When low-quality or misleading information floods the system, it becomes harder for anyone to extract real value. He linked this concern to two forward-looking signals: the approach of the final years of the 2030 Agenda, which depends heavily on data for evidence-based policy, and the inclusion of “information integrity” in the political outcome of COP30 for the first time. For Canela, 2026 will test whether these commitments translate into concrete action.

5) Responsible AI: transparency without accountability

Susan Ariel Aaronson finally emphasized that debates around responsible AI are likely to sharpen in 2026, particularly around enforcement and accountability.

She argued that many corporate commitments to responsible AI are inconsistent in practice. Drawing on her own research, Aaronson described comparisons between what companies claim in technical documentation and public statements and how their chatbots actually behave over time, especially on issues related to user welfare (such as responses to self-harm) and user rights (such as data protection and freedom of expression). While some systems improved, others showed gaps or reversals, suggesting that voluntary commitments alone may be insufficient.

Aaronson also highlighted the challenges facing the Independent Evaluators Forum, an emerging effort to assess AI systems. The forum’s central problem, she noted, is limited access to training data and evaluation inputs, which makes credible, independent assessment difficult. Without some form of mandated disclosure or cooperation, she argued, transparency cannot easily translate into accountability.

More broadly, Aaronson warned that a public “wake-up on data” has yet to occur. Even where laws exist, such as data protection rules, enforcement gaps persist, including around practices like web scraping. For her, 2026 will hinge on whether governance mechanisms move beyond aspirational transparency toward enforceable accountability.

Inclusion beyond access: language, participation, and the digital divide

The conversation also returned to the unresolved challenge of inclusion. Canela underscored that the divide is not only about the 2.6 billion people still offline, but also about meaningful participation for those who are connected. He highlighted the scale of linguistic exclusion: with roughly 8,000 languages spoken globally, only a fraction can meaningfully interact with digital ecosystems in their own language — suggesting a much deeper participation gap than connectivity statistics alone capture.

Closing reflections: three words for 2026

In the final minutes, Verhulst asked each panelist to name one concept likely to define the year:

• “New data institutions” (Sundwall)
• “Cooperate or isolationism” (Canela)
• “Manipulation” — specifically “neural manipulation” (Aaronson), including concerns about AI agents influencing users in ways people may not recognize

Key Takeaways

Across perspectives, the webinar surfaced a shared set of tensions that will likely intensify in 2026:

• Data governance is increasingly shaped by actors beyond the state, raising the need for new institutional models.
• “Open” needs to be treated as a choice within governance design, not a substitute for strategy, sustainability, or ethics.
• Quality, accountability, and information integrity are becoming central constraints — especially in an environment where traditional accountability institutions face pressure.
• Inclusion requires more than access: language, context, and meaningful participation must be part of the governance agenda.

As Verhulst noted in closing, the discussion was intended not as a final word, but as the basis for a research agenda — and an action agenda — for the year ahead.

Looking Back to the Future: What’s Next for Data & AI in 2026? was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

In 2025, the use of generative AI continued to grow, both fueling and being fueled by open data. We saw changes to rules and standards around data and AI governance. We saw shifts in public attitude and growing concerns about the data winter.

With coming changes to landmark AI laws, the economy, and communities themselves, 2026 could be a year of even more profound change. Given this, there is an urgent need to take stock of the current situation and the emerging opportunities that data practitioners should be aware of.

On January 8, 2026, The GovLab will host a small group of thinkers across regions and sectors to exchange insights on what will shape the coming year. The international, cross-sector discussion will include presentations from:

• Stefaan Verhulst, Co-Founder and Chief Research and Development Officer, The GovLab (Moderator)
• Jed Sundwall, Executive Director, Radiant Earth
• Guilherme Canela De Souza Godoi, Director for Division for Digital Inclusion and Policies and Digital Transformation (CI/DPT) and Secretary of the Information for All Programme, UNESCO
• Susan Ariel Aaronson, Research Professor of International Affairs and Director, Digital Trade and Data Governance Hub. Elliot School of International Affairs, George Washington University

This event will be a 60-minute webinar, composed of several, brief presentations each surveying a different topic or area of interest. Each presentation will be 15 minutes and be followed by a 5 minute question-and-answer period.

Register now here.

NEW WEBINAR: 2025 → 2026: Looking Back to the Future. What’s Next for Data & AI in the Coming Year? was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

In 2025, the use of generative AI continued to accelerate, even as open data initiatives, and data access policies began to slow; signaling what we have started to describe as an emerging data winter. This growing imbalance underscored a central challenge of the AI era: while demand for data is expanding rapidly, the governance models that once enabled access, trust, and reuse are no longer keeping pace.

Against this backdrop, The GovLab’s Data Program focused on re-imagining data governance for an AI-driven world — one that moves beyond narrow notions of openness toward data access that is systematic, sustainable, and responsible by design. Throughout the year, we worked across a wide range of domains — from humanitarian response and health and wellbeing to local decision-making — testing how new governance approaches can enable data reuse while preserving legitimacy, accountability, and public value.

2025 was therefore both a busy and a reflective year. Alongside major initiatives exploring the opportunities and risks of generative AI, we also critically examined long-standing assumptions about data sharing, openness, and access. In doing so, our work increasingly centered on the institutional, social, and governance conditions — such as stewardship, purpose-driven access, and social license — that will be required to ensure data ecosystems remain viable and trustworthy in the age of AI.

In this blog, we summarize some of that work. We discuss our work with a host of partners ranging from Microsoft, UNICEF, UNESCO and UNHCR, the Wellcome Trust, Siegel Family Endowment, Henry Luce Foundation, Kluz Ventures, the Institutional Data Initiative at Harvard University’s Law Library, the French Development Agency, and many others. We highlight four major areas of practice that guided much of our efforts and describe some projects indicative of that work.

In this way, we hope to both take stock of our achievements and describe the foundation we hope to build on for the forthcoming year. We hope this collection is as useful to you as it is to us and that it can help ring out 2025.

Improving Access to Non-Traditional Data for the Public Good

Throughout the year, The GovLab continued its research on non-traditional data sources — digitally captured, mediated, or observed data such as mobile phone records, online transactions, or satellite imagery — and the ways it is reshaping how we identify, understand, and respond to public interest challenges. This work builds on the research we conducted as part of the Third Wave of Open Data where we argued that privately held datasets, responsibly reused and harnessed through cross-sector collaboration, could generate public value at scale.

A cornerstone of our work this year was our collaboration with the Wellcome Trust’s Discovery Research Programme — Social Data for Health: Advancing Health and Wellbeing Research Across Disciplines. Written by Hannah Chafetz, Adam Zable, Sara Marcucci, Christopher Rosselot, and Stefaan Verhulst, it offers an in-depth study exploring how non-traditional data is being used in health and wellbeing research around the world — with a focus on the United Kingdom and low- and middle-income countries — and detailed reference guide. Drawing on over 290 studies, 5 interactive studios, numerous interviews, and insights from 23 expert reviewers, we found that social data is being used to:

• Enhance early detection of health risks and outcomes
• Reveal behavioral and environmental determinants of wellbeing
• Illuminate structural inequities and systemic health drivers
• Power cross-disciplinary research at the intersection of technology, ethics, and society

We supplemented this report with a variety of blogs, articles, and other pieces of analysis. “Monitoring the Re-Use and Impact of Non-Traditional Data” and “Unlocking Public Value with Non-Traditional Data: Recent Use Cases and Emerging Trends, published on the Open Data Policy Lab, both highlighted innovative examples of how organizations around the world used non-traditional data sources. They noted that while many examples tend to center on research (particularly health research), NTD can fill gaps in official data.

Additionally, with the support of The Second Lancet Commission on Adolescent Health and Wellbeing and UNICEF, we conducted the Youth Solutions Labs. The Youth Solutions Labs were a series of remote co-design workshops with over 120 young people from around the world. Among other topics, we sought to understand how young people feel about the re-use of data — including non-traditional data — as it relates to critical issues impacting their health and wellbeing. Our report, Designing Shared Data Futures: Engaging Young People on How to Re-use Data Responsibly for Health and Well-being, provides several insights from these discussions and the types of non-traditional data that may be harnessed.

Separately, Stefaan Verhulst contributed to an editorial in the open source journal Data & Policy on “Anticipating human mobility: Methods, data, and policy in forecasting and foresight”. The article describes the methodological frontier of forecasting and foresight and the ways that mobility datasets can contribute to that. It looks at the ethical and governance challenges of new data sources, the need for methodological pluralism, and the importance of ethical frameworks.

Advancing Data Commons Infrastructure for Responsible AI

In 2023 and 2024, the topic of generative AI rose to the forefront of the public consciousness. Conversations emerged about the opportunities and risks they presented, the rules that might govern their development and application, and what was required for responsible, effective use. The GovLab led the way in some of this work, arguing in A Fourth Wave of Open Data? Exploring the Spectrum of Scenarios for Open Data and Generative AI that generative AI offered significant potential for the open data movement, particularly when combined with data commons — collaboratively governed ecosystems that pool diverse datasets with shared standards, enabling responsible access for public benefit.

In 2025, we built on that foundation. We not only continued our knowledge leadership — developing repositories on data commons and generative AI applications, publishing a blueprint demonstrating to data stewards how they could create data commons for public-interest AI uses — but we also put our theory into practice with the New Commons Challenge.

Launched in March, The New Commons Challenge was an innovation challenge that offered two $100,000 grants to global changemakers who could provide compelling examples of how data commons might serve the public interest and unlock AI’s potential to solve complex challenges. The challenge received hundreds of applicants, which our team and an international panel of experts reviewed. In October, we selected two winners — the Amazon Rainforest Evolution Index and Malawi Voice Data Commons — and held a ceremony in New York.

In this way, we hope to show that fostering a responsible data ecosystem requires action. We will spend the next year guiding our winners and reporting back on their progress. We will also continue looking at ways we can incubate additional proposals to success. We aim to have more details on this work in the new year.

Advancing Digital Self Determination

A major gap in much of the current work on data and AI is not only trust, but rethinking agency and self-determination. While technology advocates often emphasize the benefits of data-driven tools, many individuals and communities lack meaningful ways to understand, shape, or contest how these systems affect their lives. As a result, the issue is not simply skepticism, but a structural absence of voice, choice, and agency.

When data and AI systems are deployed without mechanisms that enable people to express preferences, set boundaries, or influence governance, data use is experienced as extractive rather than empowering. Openness, rather than being a public good, can come to feel weaponized — serving institutional or commercial interests while bypassing those from whom data originate or whom AI systems most impact. This erodes public support not because people are anti-innovation, but because innovation is perceived as happening to them, not with them.

Framed this way, data sovereignty alone is insufficient. While it focuses on who controls data at an institutional or jurisdictional level, it often leaves unanswered questions about how individuals and communities exercise agency within those structures. Digital self-determination complements sovereignty by emphasizing the capacity of people and groups to meaningfully participate in decisions about data reuse, AI deployment, and value distribution — across the entire data lifecycle.

Without embedding agency into governance — through participatory processes, social license mechanisms, and accountable intermediaries — efforts to scale AI will continue to face legitimacy deficits. Reimagining data governance in the age of AI therefore requires shifting from a model of permission and compliance toward one of collective agency, negotiated use, and ongoing legitimacy.

In 2025, we looked at various ways to advance and operationalize digital self determination. With the French Development Agency, we wrote Reimagining Data Governance: Operationalizing DSD through Social Licensing. It analyzes limitations of traditional consent frameworks and outlines a structured approach — establishing community preferences, documenting conditions for reuse, and identifying enforcement pathways — to support more participatory and trusted data governance. The report also identifies key steps needed to scale social licensing, including real-world pilots, capacity building, improved oversight mechanisms, and stronger institutional support.

We also began work with several partners in Switzerland focused on digital self-determination and social license. A component of this was our participation in the Swiss Data Space Forum in Rotkreuz. Through a session on “Digital Self-Determination: Unlocking Data with Agency”, we introduced core questions about how DSD principles can move from high-level concepts to actionable design and governance features, especially in contexts where concerns about data extraction, AI-driven demand for data, and limits of consent frameworks are increasing. Through two workshops convened with partners, discussions focused on how individuals and communities can articulate, document, and enforce their data-use preferences, and how these expectations can be embedded into governance models, data-sharing agreements, and accountability mechanisms.

Developing a New Science of Questions

A major gap in much of the current work on data and AI is not only one of trust or public support, but of misaligned demand. Data access, openness, and AI development continue to be largely supply-driven — shaped by what data are available, what technologies are possible, or what institutions are willing to share — rather than by clearly articulated societal priorities and public needs.

Making data access more meaningful, legitimate, and effective therefore requires renewed attention to the science and practice of asking questions. Without well-defined questions — grounded in lived experience, policy priorities, and public interest goals — data reuse risks becoming extractive, performative, or disconnected from real-world problems. When questions are poorly specified or absent, openness can feel abstract, and AI systems appear to optimize for technical feasibility rather than social value.

With support of the Henry Luce Foundation, we convened the International Committee on Questions which led to a new concept of the Q-Lab — a new venture to develop a knowledge center devoted to developing well-defined questions. We developed several knowledge products as part of this work.

Inquiry as Infrastructure is a paper that explores what constitutes a good question — one that is not only technically feasible but also ethically grounded, socially legitimate, and aligned with real-world needs. Beyond Answers Presented by AI, meanwhile, looks at the limitations of AI systems, the roles questions can play within societies, and proposes the concept of The QLab. Stefaan Verhulst further expanded on these arguments with his blog “Generative AI and the New Tabula Rasa”, an examination of the “paradox of AI” and how it can become not merely a collaborator but a crutch.

We also worked with the Siegel Family Endowment to expand our DATA4Philanthropy initiative with new resources on the role of questions in philanthropic grant making processes. We published new blogs on questions-driven methods and a case study detailing lessons learned from Siegel’s methodology.

Beyond this work, The GovLab built on its 100 Questions Initiative — an effort to map the world’s 100 most pressing, high-impact questions that could be answered if relevant datasets were leveraged in a responsible manner. In 2025, the initiative expanded its reach with the launch of a new domain on Women’s Health Innovation, developed in partnership with the Centre for European Policy Studies (CEPS) under the Gates-funded R&I project. To surface the most pressing, data-actionable questions in this field, we created a topic map, narrative, and convened a global cohort of 77 “bilinguals” — practitioners, researchers, policymakers, clinicians, innovators, and community leaders working at the intersection of women’s health and data — from more than 30 countries.

This topic map was then refined through an iterative process of consulting with data bilinguals and public voting, leading to 10 urgent questions on women’s health. These 10 questions offer a clear, demand-driven agenda for the future of women’s health innovation. We aim to expand on this work in the new year and see how this agenda can be meaningfully acted upon.

Cultivating a Community of Data Stewards

Finally, The GovLab continued its years-long effort to foster an international, multidisciplinary cohort of data stewards, responsible data leaders empowered to exchange data for the public benefit.

In May, we worked with The Data Tank, Swiss Data Alliance, Opendata.ch, and Hasler Stifung to host the Data Stewards Intensive Course in Switzerland, an exercise that brought together 21 participants from federal agencies, cantonal authorities, the private sector, and civil society for a five-day immersion into responsible data reuse and data stewardship. The program combined expert-led sessions, field visits, and hands-on exercises focused on developing the human and institutional capacities needed to advance data collaboratives and data spaces.

In July, The GovLab worked with the Maryland Department of Information Technology to lead a multi-day Data Stewardship Bootcamp for Maryland’s Chief Data Officers, aimed at strengthening responsible data use and re-use across the state government. The event brought in guest speakers from across the region and challenged chief data officers through simulations, exercises, and more.

A final major highlight of the year was our Fourth Annual State of Open Data Policy Summit, which brought together global policymakers, researchers, and practitioners to examine how generative AI is reshaping the open data landscape. In his opening remarks, Stefaan Verhulst outlined the emergence of a “Fourth Wave of Open Data,” emphasizing the dual reality of AI’s potential to democratize access alongside growing risks of a “data winter” as governments and platforms restrict key datasets.

***

The work above only represents a fraction of our total output. Throughout the year, we participated in countless events, produced dozens of blogs, engaged in interviews and podcasts, and relaunched various websites (a full listing of which can be found here). Still, the above represents a useful cross-section of our contributions to the field and a collection of the impactful work we hope to build on for the following year — particularly the role that data commons can play in securing community agency and control and reinforcing responsible models for generative AI development.

We at The GovLab wish you a happy holiday season and look forward to engaging you all in the New Year. If you’d like to work with us on the work listed or any other projects, please don’t hesitate to reach out at datastewards@thegovlab.org.

Reimagining Data Governance in the Age of AI: Our 2025 Year in Review was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

FOR IMMEDIATE RELEASE

New York, September 25, 2025 — Artificial Intelligence is only as good as the data it draws upon. For public-interest uses — from emergency response to localized decision making — high-quality, representative, and responsibly governed datasets are essential. Yet too often, the data needed for humanitarian interventions or local problem-solving is fragmented, inaccessible, or not aligned with community needs and expectations.

Data commons address this gap. By pooling datasets under clear governance rules — with shared purpose, stewardship, and safeguards — they create trusted infrastructure for responsible AI. Unlike ad-hoc data projects, commons are designed for reuse, interoperability, and accountability, ensuring that AI solutions are not just technically sound but also socially licensed.

“The future of AI in the public interest depends on more than algorithms — it depends on the data we choose to share, how we govern that data, and whether communities see value in its reuse.” — Stefaan Verhulst, The GovLab

“From disaster relief to public health, the fastest path to impact is making frontline data findable, trustworthy, and reusable.” — Andrew Schroeder, VP of Research & Analysis, Direct Relief

That is precisely the objective of the New Commons Challenge: to accelerate the creation of data commons that fuel AI-driven solutions while respecting community expectations and maximizing public value.

The Challenge awards two initiatives with $100,000 each to develop data commons that fuel responsible AI innovation as it relates to disaster response and localized decision making. It is an initiative by the Open Data Policy Lab (a collaboration between Microsoft and The GovLab) and was conducted in partnership with Direct Relief/CrisisReady, the Institutional Data Initiative at the Harvard Law School Library and UNESCO (international observer).

On Thursday, September 25th at an event in New York, the winners of the 2025 New Commons Challenge were announced.

Certi Amazônia Institute and NYU Peace Research and Education Program have been named as the winners of the New Commons Challenge:

CERTI Amazônia Institute received the award for Enhancement of a Data Commons. Their Amazon Rainforest Evolution Index transforms environmental data into AI-ready tools to track deforestation and land-use change in Brazil’s Legal Amazon — a region covering nine states and roughly 61% of Brazil’s territory, empowering sustainable development decisions.

“The Amazon is changing before our eyes. With the Amazon Rainforest Evolution Index, we’re transforming disparate data into a shared commons that communities, scientists, and policymakers can all use. Our goal is to empower sustainable choices that protect both the forest and the people who depend on it.” — Executive Director, CERTI Amazônia Institute

NYU Peace Research and Education Program is the award winner of the Development of a Data Commons category. Their initiative, Malawi Voice Data Commons, developed in collaboration with Ushahidi, UNDP, and Mozilla Foundation, enables rural Malawians to report emergencies in native languages, creating multilingual, AI-ready datasets for humanitarian response and language preservation. The pilot will take place in Malawi with plans to scale across Sub-Saharan Africa.

“When floods approach or violence erupts, language barriers can be deadly. The Malawi Voice Data Commons allows rural Malawians to report emergencies in their own languages while preserving cultural heritage. By making these voices AI-ready, we are ensuring that humanitarian response is faster and more inclusive.” — Katerina Siira and Marine Collins Ragnet Co-Program Leads, NYU Peace Research and Education Program

The judges, composed of diverse experts across technology, civil society, academia, and international organizations also provided four honorary distinctions. These included:

Development of a Data Commons:

• Open Knowledge Brazil’s Querido Diário: Open Knowledge Brasil’s initiative converts municipal gazettes into AI-readable datasets, advancing transparency, civic oversight, and inclusive governance across Brazil.
• Slum Dwellers International’s Know Your City Academy: Slum Dwellers International and Arkology Studio are building a data commons for 5,000+ informal settlements in 18 African countries, integrating oral histories and structured data to promote peer learning and inclusive data practices.

Enhancement of a Data Commons

• PLACE Foundation’s PLACE Hub in Nigeria: Funded by ThisisPLACE, PLACE is replicating its proven model in other regions to Abuja, Nigeria to expand its community-driven spatial data infrastructure to enable responsible AI for urban planning, disaster response, and infrastructure development in Nigeria.
• Global Strategic Litigation Council’s Advancing Climate Justice: The Climate Mobility Case Database: Tens of millions are displaced annually by the climate crisis. The Climate Mobility Case Database uses AI to distill complex global case law on climate displacement into accessible, multilingual knowledge. By distilling lessons learned from courts across the world, this open platform equips communities, practitioners, and policymakers with the tools to defend rights, strengthen advocacy, and drive policy reform for at-risk and climate-displaced people worldwide. Developed by the Global Strategic Litigation Council, in collaboration with Earth Refuge, the Raoul Wallenberg Institute, and the Zolberg Institute.

What’s Next

The New Commons Challenge is not only about recognizing innovators — it is also about building a broader ecosystem for data commons. The next phase will include:

• Incubator Support: Providing the winning teams with mentorship, technical assistance, and peer learning opportunities to refine and scale their data commons.
• Repository of Examples: Expanding our open, living library of data commons projects worldwide, offering policymakers, practitioners, and funders a resource to learn from proven models.
• Blueprint for Developing a Data Commons: Operationalizing practical guidance on how to design, govern, and sustain data commons in ways that align with community expectations and serve the public interest.

Together, these efforts will ensure that the Challenge goes beyond awards — laying the foundation for a global movement of data commons that unlock responsible AI for societal good.

For more information about the New Commons Challenge and the 2025 awardees, visit newcommons.ai or email us at newcommons@opendatapolicylab.org.

Building the Future of Public-Interest AI: Certi Amazônia Institute and NYU Peace Research… was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

How can we build and sustain data commons that balance openness, and trust, while fueling innovation for the public good? Can they ensure that communities and other networks have a say in how their data is used? Can they ensure that public interest organizations can get access to the data they need to meet societal challenges?

In “Data Commons for the Public Good,” the Open Data Policy Lab’s fourth panel on the Fourth Wave of Open Data, Vivienne Ming (Chief Scientist, The Human Trust), Angie Raymond (Director of Data Management and Information Governance, Ostrom Workshop), Heather Coates (Data Steward & Data Librarian, Indiana University), and Alek Tarkowski (Director of Strategy, Open Future) joined The GovLab’s Stefaan Verhulst to answer these questions.

Together, they discussed design principles, co-creation, and purpose of data commons.

Design Principles and Incentives

The discussion began by asking participants about the foundational aspects of data commons arrangements. What principles guide data commons? How is understanding and trust created among those that provide and use data within the commons?

Vivienne Ming responded to the questions by explaining design principles guide The Human Trust’s use of unstructured human development data to build biobehavioral foundation models.

“Instead of predicting the next word in a sentence, we want to predict the next beat in a lifetime,” Ming said.

According to Ming, the Trust’s data-pooling model removes data ownership as the factor that complicates and slows down innovation. Developers using the Trust’s data should be able to touch the entire dataset, generate new data, and add value back in.

The Trust depends on individuals providing personal and sensitive data including medical, education, and legal records. Ming made the case that both individuals providing data and groups needing access to data stand to benefit from a data commons approach. Data contributors receive wellness recommendations while entrepreneurs and researchers get to interact with large datasets to which they would not otherwise have. Access at large scale is not often accessible given the proprietary nature of traditional, privately-held governance models.

Angie Raymond from the Ostrom Workshop pointed the audience to a December 2021 article on how to apply Elinor Ostrom’s eight design principles to data commons governance:

• Clearly-defined boundaries;
• Appropriate rules;
• Collective choice;
• Monitoring;
• Sanctions;
• Conflict resolution mechanisms;
• Right to self-governance; and
• Interoperability.

Raymond distinguished between data commons and trusts, referencing that different types of collaborative data governance models (e.g. commons, trusts) are needed based on the relative sensitivity of data held within such models. In short, different approaches benefit different types of work.

Problem Orientation and Co-Creation

The participants then discussed how data commons facilitate cooperation. They noted that the data commons governance model exists to drive responsible use of open data in solving public problems. Commons should be problem-oriented and co-created.

In the context of academic research, Heater Coates from Indiana University referenced the need to shift mindsets from extracting data and delivering solutions to prototyping tools that then allow the community to co-create data and data applications.

“How do we begin to work with the community and not study them?” Coates said.

Vivienne Ming agreed, saying, “To me, it’s like hammers for good. Don’t hit people with hammers; build things. We can get caught up in building these frameworks or these philosophical commitments when the starting point for doing good with AI is problem-solving and problem-solving means owning the whole problem.”

As institutions, such as research bodies at universities, must contract with entities and not individuals, a crucial step in co-creation is recognizing data providers as communities and not just individuals.

Alek Tarkowski continued the argument for protecting collective data rights, particularly in the context of linguistic data, by pointing to the Bloom GenAI model from BigScience in 2022. The data commons governance model holds promise for GenAI applications when considered as a framework for data collection.

Future and Purpose of Data Commons

Data collaboration for AI-usable data is still in its infancy. Experimentation is needed to determine what governance models are fit for which purposes. This teed up Stefaan Verhulst’s final question to the panelists: How is purpose defined in a commons-focused manner, and what questions should be prioritized?

For the Human Trust, Vivienne stated, the board simply asks, “Does this project (that uses Human Trust data) advance Human Development?” If so, the board grants approval.

Angie Raymond remarked that amid a data explosion (and the potential for a data winter), granularity on the types of data is essential. Heather Coates added that the question space originates in relationships and co-creation.

Alek Tarkowski closed out the seminar by pointing to alignment assemblies as a vehicle for using collective intelligence to define the “good” that AI can help create.

***

These are just a few of the reflections offered in our latest seminar on the Fourth Wave of Open Data. To follow the full discussion, watch the video here.

Stay tuned for the announcement of our final seminar and our forthcoming State of Open Data Policy Summit, which will examine larger challenges and trends.

Fourth Wave of Open Data Seminar: Data Commons for the Public Good was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

With over 170 applicants spread across six continents and countless sectors, it’s clear that there is an urgent need for resources to develop data commons.

Last week, we closed applications for our New Commons Challenge — an open innovation challenge seeking to foster the use of data commons for the development of AI for local decision-making and humanitarian response.

The response to the New Commons Challenge has been a powerful signal of the growing interest in data commons as a catalyst for public interest AI and locally grounded impact. Between April and June, we received 173 submissions from around the world.

Although the Challenge focused on two priority areas — humanitarian response and local decision-making — the proposals spanned a wide range of data types and use cases.

As we move into Phase Two, we are leveraging the current review process as a listening exercise — helping us gauge the appetite for and potential of data commons around the world. What kinds of initiatives are resonating most with AI, data and other practitioners? Where is this momentum coming from — and what can it tell us about the emerging future of collaborative data governance?

Geography

Of the 173 submissions, 154 are eligible for the current Phase One review. These entries span 51 countries from across the world and reflect the widespread global interest in data commons initiatives.

Award Type

Submissions fall into two categories: 103 entries (67%) for Award 1, focusing on creating a new data commons, and 51 entries (33%) for Award 2, focusing on enhancing an existing data commons.

All applicants have to be affiliated with an established organization, and submissions primarily come from non-profits, start-ups, and research organizations (universities or research institutions).

Applicants

Data and Submission Type

Nearly all submissions involve multiple types of data, but the most common dataset types included:

• Text or unstructured data (language and linguistics, legal data, research and education, news and media, biomedical data)
• Structured data (e.g. national statistics, sensor networks, administrative data)
• Imagery (e.g. knowledge graphs, spatial images, artwork, data visualizations, geospatial datasets)

Many submissions also incorporated audio and video data, signaling a clear broadening of the types of data being considered for use in data commons. This trend reflects a growing recognition that multimodal data can offer valuable insights, especially when designing AI systems responsive to local needs and contexts.

The New Commons Challenge is specifically interested in data commons that support disaster response or local decision-making. Around 75 submissions focused exclusively on supporting local decision making. Of the 69 total submissions focused on humanitarian response, the vast majority also discussed how their data commons would support local decision making in crisis response or prevention.

Next Steps: Phase Two

After we conclude our review of Phase One concept notes, selected applicants will be invited to submit a full proposal outlining their use case and implementation. An interdisciplinary group will assess proposals for feasibility, innovation, and impact potential.

Of the shortlisted applications, two teams will each receive $100,000 in funding, along with mentorship, networking opportunities, and global exposure to support the execution of their data commons.

Decisions will be finalized in the coming months. To learn more about The New Commons Challenge, you can visit newcommons.ai.

Signals of Demand: What the New Commons Challenge Tells Us About the Need and Opportunity of Data… was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.

This blog is part of a continuing series on the Fourth Wave of Open Data. To read our first blog in the series, click the link here.

Over the last three months, the Open Data Policy Lab has explored how the Fourth Wave of Open Data fundamentally changes the way we conceive of the open data movement.

In the first episode, we spoke with leading data practitioners about how the combination of open data and generative AI offers significant potential. In the second episode, we discussed how open data might be made conversational through the application of AI tools.

On 5 May, the discussion looked at something more foundational: How do we make data ready for AI? How do we future-proof open data?

Joined by an international panel of experts that included Stefan Baack (Mozilla Foundation), Zhamak Dehghani (Nextdata), and Chris Shaffer (BrightQuery), Stefaan sought to understand what AI readiness for open data means and what strategies we might pursue regarding data curation, standardization, and quality.

Research on Open Datasets for LLM Training

The conversation opened with a brief discussion with Stefan Baack on recent research from the Mozilla Foundation and EleutherAI that examined best practices for creating open datasets for LLM training.

As he noted, the piece was the result of a short workshop, one that asked the question, “Is copyrighted data truly essential to build AI data or can we rely on openly licensed datasets to make AI more equitable?”

The results, which included the inputs of 30 AI scholars and practitioners, found significant opportunities to build open, transparent datasets for AI development.Acting on this potential meant overcoming certain challenges — principal among them scalability.

“I would summarize all the challenges that were identified as a lack of scalable solutions to identify and acquire this kind of data, scalable is really the key word here,” said Stefan.

He noted that there was no shortage of permissibly licensed material. Much of this work existed in the public domain due to a failure to renew copyright. However, much of this content has never been digitized or has poor Optimal Character Recognition (OCR), which is needed to train data.

Still other material existed on HTML pages with non-standard terms of use or with terms that required a human operator to understand. A page with a Creative Commons license might only intend part of the page to be in the commons, not everything in its entirety.

The paper identified possible solutions to address these problems. The open data movement could develop technical standards to encode preferences throughout the web on how it wants data to be used and not used (e.g. explicitly indicating what content should not be used for AI training). There could also be greater effort around making terms of use more machine readable.

What Stefan closed on, though, was a call to build a community of AI curators around shared interest, curators who could better define these conversations and identify the specific solutions needed for their domain.

“I think there needs to be a community made up of both AI builders and open data curators around shared interest. I think it is important that the AI community is not just a consumer of open data, but also in various forms, a contributor to it,” he said. “I think the Open Data community could be really important in developing [technical] standards and adopting them at scale.”

Federated Models of Data Ownership

Following these remarks, Stefaan turned to Zhamak Dehghani. Based on her experiences, he sought to understand how she saw data evolving. How did changes in how we handle data affect its readiness for AI?

Zhamak responded by noting that the challenges raised around data sharing, discoverability, and scalability had been a major focus of her work. At NextData, the solution they had settled on was one built around a federated model of data ownership.

“The reason for that is the problems that Stefan mentioned around discoverability, around understanding what the data is about and then how it can be used. […] We’re really looking at this intelligent block that can describe itself. It has all the metadata to describe itself. It can computationally govern whether this particular use case that is trying to access my underlying data ultimately should or should not be used.”

“At NextData, we’re trying to solve this problem of data sharing at scale with trust and product thinking built into it with a technology that abstracts the complexity that sits beneath this concept of autonomous data as a product,” she continued.

Stefaan followed up by asking if there were any ways we could accelerate this approach and what kinds of investments that would take.

Zhamak responded by noting that standards always accelerate innovation. While they’re hard to get because they require people agreeing with one another, a few standards would at least start a conversation and allow for people to adapt that to their experiences.

“We need the hourglass of data innovation, a narrow waist with a small set of standards that, frankly, right now are missing, including those around access control.”

Building New AI

Finally, Stefaan turned to Chris of BrightQuery. Noting that the organization had significant expertise working with statistical agency, including on opendata.org, Stefaan wanted to know what challenges they saw emerging in the field and possible approaches to make data AI ready.

Chris, in turn, drew on his recent project experiences. He spoke on his work with the US government to evaluate its data, build a government data chatbot, and help establish an open entity dataset.

The question of how to make open data AI-ready was one that came up very quickly in BrightQuery’s effort to evaluate government data. Namely, what does “AI-ready” mean?

There’s AI ready for training an LLM, AI-ready for building a commercial chatbot that will provide the right answer, AI-ready for answering questions about structured data, and AI-ready for predictive analytics, among other applications.

BrightQuery thought about what the commonality between all these possible applications might be and settled on issues of accessibility and interoperability.

“Of the things that we found is almost universal is we have to look at the difficulty of just crawling a website, scraping data off of it, pulling data from pulling information from OCR. And this is all data that is intended to be open. This is data that is published by government agency that’s intended to be as easy to work with as possible.”

He also mentioned the importance of data provenance for building systems, noting that, when it comes to accurate AI results, having well-bounded data whose origin can be identified is critical.

“It’s not only important enough to not abuse work that people put out into the wild. It’s also important just to get the answers right. You’ve got to know where every number is coming from. Not knowing if your financials might be coming from Reddit or a sci-fi book is terrifying.”

***

These are just a few of the reflections offered in our first seminar on the Fourth Wave of Open Data. To follow the full discussion, watch the video here.

Stay tuned for our reflections on our next seminar “Data Commons for the Public Good”.

Fourth Wave of Open Data Seminar: Future-Proofing Open Data: Making Data AI-Ready was originally published in Data Stewards Network on Medium, where people are continuing the conversation by highlighting and responding to this story.