Generative AI in the Real World: Stefania Druga on Designing for the Next Generation

Why It Matters

Empowering young learners with agency‑driven, multimodal AI transforms education and drives a market shift toward collaborative, on‑device intelligent tools.

Key Takeaways

• Kids prioritize playful, non‑utilitarian AI uses over efficiency.
• Provide AI “agency knobs” to balance assistance and user control.
• Socratic co‑pilot asks questions, fostering computational thinking and debugging.
• Multimodal, on‑device models enable real‑time math feedback on phones.
• Open‑source small models (Gemma, Llama, Mistral) power low‑latency apps.

Summary

Stefania Druga, former DeepMind research scientist, explains how generative AI must be re‑engineered for the next generation of users—primarily Gen Z learners. She contrasts the typical utilitarian, task‑automation mindset of adult‑focused tools with the playful, exploratory ways children experiment, such as training models to spot quirky hairlines or generate backhanded compliments. Her research shows that young users need an "agency knob" that lets them decide how much the AI should assist versus challenge them. By embedding a Socratic co‑pilot into the Cognimates platform, the system asks guiding questions, offers affirmations, and only intervenes after repeated struggles, effectively teaching computational thinking and debugging skills. Examples include the co‑pilot prompting, "Which way do you want it to move?" instead of inserting code blocks, and the Math Mind app that scans handwritten work, detects algebraic misconceptions, and asks probing questions in real time. Studies across eleven countries revealed that such affirmations doubled coding time and that multimodal, on‑device models could deliver instant feedback without cloud latency. The broader implication is a shift toward AI that collaborates, not delegates. Designers must build tools with adjustable agency, multimodal capabilities, and low‑latency open‑source models to support personalized, inquiry‑based learning, reshaping both education and consumer AI products.

Original Description

How do you teach kids to use and build with AI? That’s what Stefania Druga works on. It’s important to be sensitive to their creativity, sense of fun, and desire to learn. When designing for kids, it’s important to design with them, not just for them. That’s a lesson that has important implications for adults, too. Join Stefania and Ben to hear about AI for kids and discover what it reveals about AI for adults.

Follow O'Reilly on:

LinkedIn: https://www.linkedin.com/company/oreilly/

Facebook: http://facebook.com/OReilly

Instagram: https://www.instagram.com/oreillymedia

BlueSky: https://bsky.app/profile/oreilly.bsky.social