On 4 September 2025, the Massachusetts Institute of Technology hosted an online seminar—part of the AHA Speaker Series—organized by the MIT Media Lab’s Advancing Humans with AI research program. Over Zoom, moderator Dr. Pattie Maes welcomed attendees and introduced the panel of researchers: Ron Ivey from the Harvard Human Flourishing Program, Jonathan Teubner of Harvard’s Human Flourishing Program, Nathanael Fast of the USC Marshall School’s Neely Center for Ethical Leadership, and Ravi Iyer of USC’s Psychology of Technology Institute. Together, they examined both the benefits and risks that AI chatbots pose for children and young people, and proposed multi-level design principles to ensure AI supports human flourishing.

Opening Remarks and Scope of the Conversation

Dr. Maes began by framing the urgency of the topic: chatbots powered by large language models are increasingly accessible to minors, offering potential mental-health support even as they risk deepening social isolation. She emphasized that close social relationships are essential to human flourishing and invited the panel to explore how design choices at the technological, organizational, market, policy, and cultural levels can mitigate risks while amplifying AI’s positive impact.

Risks and Design Choices at Multiple Levels

Ron Ivey outlined the seminar’s core thesis: at every tier—from model architecture and product development to investor incentives and cultural norms—there are choices to be made that will determine whether AI strengthens or undermines children’s social connectedness. He showed evidence of a long-term decline in youth social engagement and warned that chatbots, if designed without care, could exacerbate loneliness rather than alleviate it.

Defining and Measuring Human Flourishing

Building on this foundation, Jonathan Teubner described the Human Flourishing Program’s framework, which defines flourishing across six domains—meaning and purpose, close social relations, character and virtue, happiness and life satisfaction, health, and agency. He highlighted the Global Flourishing Study, involving 240,000 participants in 22 countries, as the largest ongoing effort to track these dimensions over time. Teubner argued that any AI intervention aimed at youth must be evaluated against its impact on these core facets of well-being.

Purpose-Driven Paradigms and Impact Indices

Turning to industry engagement, Nathanael Fast introduced the Neely Center’s work on shifting AI development from a purely profit-driven paradigm to one grounded in human-centered values. He described two “design codes”—one for social media and one for AI companions—and detailed the Neely Indices, nationally representative surveys that now measure how both adults and teenagers experience social media and AI in their daily lives. Fast emphasized the value of tracking youth experiences longitudinally to inform design and policy.

Bridging AI and Human Connection Through Collaboration

Ravi Iyer closed the formal presentations by sharing his nonprofit initiative, Noasis Collaborative, which aims to bridge AI developers with social-science researchers. Supported by philanthropic networks, Noasis convenes multi-stakeholder workshops to align AI product roadmaps with insights about human connection and to secure funding for research on children’s social capabilities.

Toward a Child-Centered AI Design Paradigm

Throughout the discussion, the panel underscored five key domains for collective action: technical safeguards built into AI systems; corporate cultures and business models that prioritize child welfare; investor criteria that reward child-friendly design; policy frameworks led by G20 nations to codify international standards; and community-level education to help families and schools navigate AI responsibly. By measuring youth well-being across these axes—and by embedding safeguards into both product code and organizational incentives—the researchers presented a roadmap for AI innovation that not only avoids harm but actively fosters children’s social and emotional development.

The MIT Media Lab is a pioneering research center at the Massachusetts Institute of Technology that brings together technologists, artists, and designers to explore bold innovations at the intersection of science and creativity, driving transformative advances in robotics, human–computer interaction, and digital art worldwide.

The Conf is a platform that reports on scholarly conferences, symposia, roundtables, book talks, and other academic events. It is managed by a group of students from leading American and European universities and is published by Alma Mater Europaea University, Location Vienna.