The Next Frontier of AI Video Is Control
A16z General Partner Jennifer Li interviews fal co-founder Gorkem Yurtseven and Head of Engineering Batuhan Taskaya about the implications of real-time generative video. They explain the technical advancements behind H3 Max, fal’s post-trained version of MiniMax’s open-weight video model. The discussion covers how combining model post-training with systems and hardware optimization led to significant reductions in generation time while preserving quality. This speed has enabled experiments with continuous video, including streams that can remember previous scenes and respond to new directions. The conversation also explores the next challenge: moving beyond speed to control aspects like camera movement, lighting, characters, motion, and lip sync, and what these capabilities mean for professional creative workflows requiring predictable tools.
Original sources
Podcast The a16z Show
AI Radar wrote this summary from the publisher’s notes with AI, not from a transcript. How it works.