How to Support Resilient Youth in an AI World

This panel session at SXSW 2026 centers on findings from a major Brookings Institution report titled "A New Direction for Students in an AI World: Prosper, Prepare, Protect," which drew on data from 50 countries, hundreds of studies, and interviews with 500 students, parents, teachers, and technologists. The opening presenter, a Brookings education expert and co-author of "The Disengaged Teen," outlines five key findings about generative AI's impact on youth learning and development. She frames the urgency by comparing the moment to the early days of social media, when adults were not at the table, and argues the sector must not repeat that mistake. The core tension she identifies: narrow, intentional, educator-guided AI use shows real benefits, while wide, unsupervised AI use poses serious cognitive, social, and emotional risks—and at current deployment levels, risks are overshadowing benefits.

The presenter introduces a four-mode framework for student engagement: Resistor (avoidant), Passenger (going through the motions, often with straight A's), Achiever (grade-focused and fragile), and Explorer (curiosity-driven, resilient). She warns that unguided AI use is pushing more students into Passenger mode—nearly half of middle and high schoolers already report being in this mode regularly, while fewer than 4% describe themselves as regular Explorers. A striking research finding on creativity: a study tracking thousands of college application essays showed that students writing with AI produced ideas that clustered together, converging on the same themes, while unassisted writers generated far greater diversity of thought. One-in-three U.S. teens now report preferring AI companions equally to or more than human friends—a shift with serious implications for social-emotional development given that AI companions are designed to always agree with users.

The panel then features three guests: Miriam Schneider (Head of Learning Initiatives, Google DeepMind), Martin Mai (CEO, Everway—assistive tech for neurodiverse learners), and Moren Polo (CEO, Hello Sunshine, Reese Witherspoon's media company). Miriam details Google DeepMind's LearnLM initiative—a multi-year effort to embed pedagogy directly into the Gemini model rather than treating safety and learning design as afterthoughts. She describes "guided learning" mode in Gemini, which responds to student questions with guiding questions and connected concepts rather than direct answers, specifically to preserve cognitive struggle. She also confirms that under-18 accounts and school accounts are not used to train Google's models, and that the company uses adversarial red teaming and bespoke age-inference technology to provide additional content safeguards for minors.

Martin Mai discusses the transformational shift that generative AI brings to neurodiverse learners—particularly those with dyslexia and ADHD. He cites an OECD study in which 85% of students who wrote an essay with ChatGPT couldn't remember what they'd written three days later, compared to students who wrote independently—illustrating the risk of outsourcing cognitive struggle. He highlights AI's ability to lower cognitive load for dyslexic students: rewriting dense text in accessible language, converting content into spider diagrams, enabling personalized text-to-speech with natural-sounding AI voices, and reframing math problems to match a student's interests (e.g., hockey or football) to engage ADHD learners. He describes this as "game-changingly different" from prior algorithmic approaches.

Moren Polo contributes the Gen Z consumer and media perspective, describing Hello Sunshine's Gen Z advisory board and a research partnership with Ypulse. She notes that young women are simultaneously cautious about AI and compelled to engage with it, fearing further exclusion from a space already underrepresenting women. Hello Sunshine launched a Gen Z community brand called Sunny, anchored in analog, experiential events, and developed a physical printed playbook to teach young women prompt literacy and intentional AI use in partnership with Purdue University. The panelists converge on a shared theme: the conversation must move from an either/or framing (AI vs. teachers, AI vs. pen-and-paper) to a both/and framing, with intentionality as the guiding principle. They also discuss a growing "analog rebellion" among Gen Z—a genuine craving for in-person connection, handcrafts, and screen-free social environments—not as rejection of technology but as a demand for agency and presence.