What AI Owes Children: A New Blueprint for User-Centered Beneficial Innovation
Authors: Mathilde Cerioli, Adrien Abécassis
Abstract
This chapter advances a child-centered blueprint for beneficial AI innovation, grounded in developmental neuroscience, child psychology, and children’s rights. Drawing lessons from the unintended harms of earlier digital platforms, it argues that the failure to design technologies around children’s developmental needs has produced long-term costs that could have been avoided with earlier scientific and ethical foresight. As generative and conversational AI systems increasingly shape children’s learning, relationships, and environments, the chapter asks a fundamental question: how can AI be designed to support development rather than undermine it?
The chapter examines both the opportunities AI offers for children, including personalized learning, expanded access to education, and the realization of fundamental rights, and the risks associated with poorly aligned design, particularly cognitive and emotional overreliance and the rise of parasocial relationships with anthropomorphic AI systems. It highlights structural limitations in the current ecosystem, including the disconnect between research and product development, mismatched ethical standards between academia and industry, and regulatory frameworks that remain largely reactive to harm.
In response, the chapter presents the iRAISE Alliance as a new, integrated model for child-centered AI. By connecting research, product design, policy, and public engagement from the outset, the Alliance seeks to move beyond risk mitigation toward proactive, developmentally aligned innovation. The chapter outlines the Alliance’s multi-stakeholder architecture, its emphasis on co-design with children, and its ambition to establish shared frameworks, standards, and practices that embed children’s development and rights into AI systems from the beginning. It concludes with a call to action for policymakers, industry leaders, researchers, and civil society to collectively shape an AI future that protects, empowers, and grows with the next generation.
