Koray Kavukcuoglu on frontier models, coding agents, and building AGI

Google DeepMind SVP and Chief AI Architect Koray Kavukcuoglu reflects on the path from DeepMind's early reinforcement learning work to Gemini and what it means to build AGI at Google. Google DeepMind SVP and Chief AI Architect Koray Kavukcuoglu joins host Logan Kilpatrick to reflect on the journey from DeepMind's early reinforcement learning milestones to Gemini and what it means to build AGI at Google. Watch along and learn: * What made Gemini 3.7 Flash a breakthrough for engineers * The ambitions of the Gemini 4 pre-training run * What it takes to stay at the frontier of AI research * How co-building AGI with users shapes everything the Google DeepMind team does 00:00 Intro 00:24 From models to coding agents 03:07 Gemini 4 pre-training 04:08 Why the frontier is all that matters 06:38 Leading frontier AI at Google 08:53 Why there's no test for AGI 10:58 Early days at DeepMind 15:08 Atari vs. real-world ambiguity 17:18 20 years of deep learning 18:47 The reality of engineering hill climbs 21:05 Why Google is the place to build AGI 22:19 Parallel model development 23:34 The shift to agentic coding 24:57 What makes models more intelligent Watch on YouTube: https://www.youtube.com/watch?v=Rrr2gdbvNFU