Interview
On Gemma, AI Studio and the future of AI

Omar Sanseviero
Developer Experience Lead, Google DeepMind

Paige Bailey
Lead, Generative AI DevRel, Google DeepMind
27 minHosted by Kim IsenbergRecorded at Google I/O 2026
Watch on YouTubeAbout this conversation
At Google I/O 2026, Kim Isenberg sits down with Omar Sanseviero and Paige Bailey from Google DeepMind to talk about the future of building with AI: why Google is investing so heavily in open models, what Gemma changes for developers, why Apache 2.0 licensing matters, and how tools like Google AI Studio could make AI development accessible to far more people.
Omar leads Developer Experience at Google DeepMind, working across Gemini, Gemma, the Gemini API and Google AI Studio. Paige works on creator and developer experiences around Gemini and has seen the AI developer ecosystem from Google, GitHub and Microsoft.
Topics
- Why Gemma matters in the open model race
- What developers actually need from AI models today
- Cloud AI vs. local and on-device AI
- One giant model or many smaller specialized models?
- How far non-engineers can get with AI Studio
- Sovereign AI and the global race around open models
Kim Isenberg is co-founder and Editor-in-Chief of Superintelligence, an AI newsletter published every weekday.
Read the NewsletterMore interviews
- Kari BriskiWhy NVIDIA gives away its best models
- Ming-Yu LiuIntro to NVIDIA Cosmos
- Sven BreunerThe hidden data bottleneck in AI factories