Google's Jon Barron: video models near Hollywood quality, but controllability is 2-5 years out
rms80 · x · 2026-09-02
Google researcher Jon Barron predicts video generation models are <=1 year from the Hollywood quality bar, but reaching the level of controllability VFX artists and filmmakers demand before switching over will take 2-5 years — controllability, he argues, is the harder long-term problem than quality.
3D researcher chrisoffner3d, revisiting a 5-year prediction 2.5 years in, agrees: image quality and some control (img2video) have improved, but fine-grained control for filmmakers remains limited to camera control (e.g., Genie & Atlas), with lots of room for progress in the remaining years.
More from Multimodal
- Video generation has matured: you can now direct models instead of prompting and hoping — MilitantAI · 2026-09-03
- Video gen has moved from prompting to directing — and product UX isn't ready — Kyrannio · 2026-09-03
- World Labs unveils Atlas, a new video generation model — mildlyphd · 2026-09-03
- Runway's Solaris: an interface world model that renders apps pixel by pixel — umpherj · 2026-09-03
- VideoDeltaNet open-sources hybrid attention that speeds up MiniMax H3 video generation up to 90x — realmrfakename · 2026-09-03
- jjk-explain turns any concept into a Jujutsu Kaisen-style explainer video with one Claude Code command — teortaxesTex · 2026-09-03