As artificial intelligence systems become more reliant on data from low- and middle-income countries (LMICs), fundamental questions arise about who controls that data and who benefits from its use and reuse. In many cases, the people and communities who generate this data have little say in how it’s used, and few mechanisms for recourse when harms occur.

To address these challenges, The GovLab and Agence Française de Développement (AFD) have partnered on a new project exploring more equitable, participatory approaches to data governance in AI ecosystems. Today, we are pleased to release the outcome of that collaboration: Reimagining Data Governance for AI: Operationalizing Social Licensing for Data Reuse.

The report introduces social licensing as a practical, community-centered model for governing data reuse. In contrast to traditional frameworks based on static, binary, one-time consent, social licensing supports an ongoing process of engagement, negotiation, and accountability. It reframes data governance as a shared responsibility rather than an individual burden, enabling communities to define acceptable uses of their data and move toward ensuring those preferences are honored throughout the AI lifecycle.

Building on insights from the earlier AFD and Data Tank report Responsible Data Re-Use in Developing Countries: Social Licence Through Public Engagement, which emphasized participatory engagement, this new report takes the next step by providing a practical roadmap for implementation. It positions the establishment of preferences and expectations through community engagement as the first phase of a three-part process (illustrated in the image above) and introduces a Social Licensing Toolkit containing three tools:

• A questionnaire to facilitate structured community engagement;

• Sample legal clauses that embed community-defined preferences within enforceable agreements; and 

• Scenarios illustrating how the tools can be applied in practice

While the report focuses on the first two phases—establishing and documenting preferences—it also offers suggestions on how to approach enforcement. Strategies range from formal legal agreements and national regulations, to community-based oversight, intermediary governance models like data trusts and cooperatives, and even blockchain-based tools such as smart contracts and DAOs.

Recommendations for scaling and sustaining social licensing include:

• Piloting the model in diverse sectors to test, adapt, and refine it;

• Developing hybrid compliance mechanisms that combine legal, technical, and community oversight;

• Building capacity among governments, communities, and practitioners;

• Embedding social licensing into procurement, funding, and policy frameworks;

• Creating a Center of Excellence to support implementation and learning; and

• Advancing research to measure impact and improve the model over time.

Read the full report HERE.

Explore the prior report by AFD and the Data Tank HERE

Questions or feedback? Interested in applying social licensing in your work? Reach out to us at [email protected].

Back to the Blog