With the era of digitisation and rapid advancements in computational tools for data-driven innovation, the flow of data from individuals to organisations has increased exponentially. However, the current data economy, and data sharing practices within it, serve to embed a marked power imbalance in the knowledge, agency, and participation afforded to individuals and communities over their data.

It is in this context that data stewardship as an approach to data governance assumes significance. Data stewardship seeks to address this imbalance by distributing the power to control data and the value derived from it amongst individuals and communities. Data stewards act as intermediaries that work in the interest of individuals and communities to facilitate meaningful consent, participation, and increased decision-making power to users and data generators.

Academics and practitioners across the globe are exploring models of stewardship that can help build a more equitable data economy -one that harnesses the latent societal value of data while also safeguarding individual and community rights. In particular, the concept of data trust has emerged as a possible stewardship model to facilitate human-centric data governance. As defined by the Global Partnership for AI (GPAI), a data trust is a model of stewardship wherein “data producers pool their data (or data rights) intending to collectively negotiate terms of use with potential data users, through the oversight by independent trustees, with fiduciary duties, and within a framework of technical, legal and policy interventions that facilitate data use and provide strong safeguards against misuse.”

Naturally, building sustainable data trusts necessitates an understanding of the institutionalised regulatory and legislative frameworks they may stand on. This is particularly true when seeking to reimagine and embed existing legal mechanisms like fiduciary relationships in the novel context of data governance. To viably operationalise bottom-up data governance, there is a need for top-down state interventions to enable conducive regulatory and structural environments.

To this end, Aapti, with the support of the GPAI, is undertaking an analysis of existing legal landscapes for data sharing -mapping challenges, needs, and uncertainties in operationalizing and sustaining data trusts.

Spanning the next six months, Aapti’s research on the legislative landscape for data trusts will:

• Map existing and evolving global policy frameworks for data sharing, data protection, and data rights. This will entail a review of legislation, regulatory structures, and policy directives across various jurisdictions such as the European Union, Australia, Finland, Brazil, Kenya, India, Canada, etc.
• Identify gaps and uncertainties in current legal frameworks for how they may hasten the feasibility and sustainability of data trusts. The comparative approach will feed into our development of best practices on the different aspects of data trusts.
• Analyse legal structures across key aspects for data trusts. This analysis metric will include conceptions of fiduciary duties, data rights, mechanisms for data sharing and pooling, protections afforded to individuals and communities, and more.

Through this research, we aim to offer a detailed review of the legislative and regulatory frameworks that can support data trusts. Alongside desk research and analysis of legal frameworks and academic perspectives for data trusts, we are keen to engage civil society members and experts from Low and Middle-Income Countries to ensure equitable representation of stakeholders. Building on Aapti’s existing understanding and analyses of data stewardship, the resultant vision of data trusts and their top-down needs will incorporate an agential, community-oriented lens.

In tandem with this work, the Aapti team is also collaborating with the Open Data Institute, supported by GPAI; to explore global perspectives, knowledge, practices, and instances of data trusts for societal benefit