* A Grant DOI (digital object identifier) is a unique, open, global, persistent and machine-actionable identifier for a grant.
The Oxford Global Challenges Programme (GCP) established the Generation AI initiative in 2024, with the stated aim of ensuring AI has a positive impact on future generations. With pilot funding from the Templeton World Charity Foundation, Elevate Great, and the University of Oxford, we have built a strong foundation for this programme by mobilizing a strong community of researchers, innovators, and advocates.
Over the past year, through a comprehensive programme of work, we have identified an opportunity space for impact in supporting rigorous mechanistic, outcome-oriented studies at a speed and scale necessary to influence decisions.
There is a scarcity of high-quality evidence regarding the impact of AI on child development, making it difficult for developers to make evidence-informed design decisions. In addition, we see a significant absence of integration between child developmental research, design, and technological innovation hindering translation of research into impact. We need greater incentives for evidence-based, outcome-oriented research, better metrics to understand on-platform flourishing, and new infrastructure to leverage these.
As such, GCP and our partners propose an 18 month project to tackle this through a three-pronged approach:
Promoting a greater awareness and incentive for evidence-based, outcome-oriented studies;
Developing metrics for on-platform measurements of flourishing;
Prototyping computational infrastructure to collect and access appropriate data sets, design and evaluate holistic human-behavioral metrics, and ultimately facilitate rapid outcome-oriented trial design.
We see a strong alignment between our goals and Templeton World Charity Foundation’s commitment to do research and disseminate practices supporting character development of children in a world where learning, growth, and development are increasingly influenced by AI systems. The project will synthesize existing evidence and identify design patterns in AI systems which scaffold establishment of virtuous habits such as agency/self-authorship, critical thinking, wisdom, mutual understanding, and loving relationships as well as other beneficial outcomes in learning and development of children.