Dr. Sonia Tiwari is a children's media researcher based in the San Francisco Bay Area and Director of Research at Oki Pie Lab. She holds a PhD in Learning, Design, and Technology from Penn State University, along with an MFA in Animation and a BS in Film, credentials she first put to work as a character designer in the video game industry before turning to research on how children bond with characters and, more recently, AI.

At Oki Pie Lab, she runs weekly art workshops for children ages five to ten — more than 100 to date — using what she observes each week to test where AI helps kids create and where it gets in the way. Her research includes Tiwari, S. (2025). Designing Expressive Arts Workshops for Families Using AI Tools. International Journal of Designs for Learning, 16(2), 194–204, and Tiwari, S. Designing ethical AI characters for children's early learning experiences. AI Brain Child 1, 11 (2025) — built from analyzing 60 social media videos of 20 AI characters marketed to children as young as three. She consults with edtech founders — including Jules Stories, a UK-based screen-free storytelling companion — on whether their AI characters genuinely serve children's development, and in 2025 she was selected for the Falling Walls Foundation's Female Science Talents Intensive Track.

In this episode, Sonia talks with Svenia Busson about:

• Why visual design — not just language or policy — shapes how children bond with AI characters
• The psychology of "cute" character design, from Japan's Kawaii culture to Germany's Kindchenschema, and how it's used to mask age-inappropriate AI
• The "sticky chat patterns" she's studying with Dr. Jennifer Shen (Keene University) — and their cost to kids' ability to explore beyond one interest
• Her AI Character Assessment framework in AI Brain Child, built from analyzing characters marketed to ages three to eight
• Why safety needs three layers — design, family, and law — and why design alone can't fix a systemic problem
• The patchwork of AI regulation worldwide, from loophole-prone California state laws to stricter approaches across Europe
• Rating efforts underway, from Common Sense Media and Digital Promise to the kidSAFE Seal Program
• What Blake, Carter and Velloso's CHI 2026 research reveals about why teens build their own Character.AI companions

Sonia's recommendation for a future guest: Diana Joseph, CEO of the Corporate Accelerator Forum and an innovation and entrepreneurship strategist working at the intersection of AI and organizational innovation.

Mentioned in this episode:

Education Futures Fellowship

A six-week programme starting on September 22nd, where seven world-leading experts in developmental psychology, neuroscience, and AI child safety will share everything they know with you: educationfutures.ai/fellowship.