Learning data trusts and stewardship models are legal and technical structures where learner data is held in shared trusts with explicit governance rights, democratic decision-making processes, consent layers, and revenue-sharing mechanisms for derivative model training and data monetization. These frameworks recognize that learner data is valuable and that current models often concentrate power and benefits with platforms and institutions rather than learners themselves. By creating trusts with shared governance, these models aim to rebalance power, giving learners and their representatives meaningful control over how their data is collected, used, and monetized, while ensuring that benefits from data use—including revenue from training AI models—are shared with learners rather than extracted by platforms.
This framework addresses the power imbalance in educational data, where platforms and institutions collect and monetize learner data while learners have little control or benefit. By creating shared governance structures, these models can give learners agency over their data and ensure they benefit from its use. Researchers, data governance advocates, legal experts, and educational institutions are exploring these models, with some pilot programs already testing data trust structures.
The framework is particularly significant as educational data becomes more valuable for training AI models, where establishing fair governance and benefit-sharing could ensure that learners benefit from their data rather than being exploited. As data becomes more valuable, creating equitable data governance models could become essential. However, establishing legal structures, managing governance complexity, ensuring effective representation, and creating sustainable models remain challenges. The framework represents an important evolution in data governance, but requires continued development and legal innovation to be effective.
