Fei-Fei Li explains Atlas: new-view prediction as a primitive toward AGI
量子位 · wechat · 2026-09-05
Fresh off World Labs' launch of Atlas, billed as the first multimodal world model, Fei-Fei Li and co-founders Justin Johnson and Ben Mildenhall sat down with a16z to unpack it.
Key takeaways:
- Core primitive is new-view prediction: like next-token for LLMs and next-frame for video models, Atlas predicts new viewpoints. Inputs (images with 3D camera poses) form a "spatial context" from which it renders the world from arbitrary positions.
- Generation and reconstruction unified: one model natively handles text, images, video, 3D depth maps and camera poses; Li calls this the first unification in computer vision's 50-year history.
- Massive capture reduction: dense reconstruction once needed hundreds of photos; Atlas rebuilds scenes from 3-25 images (e.g., Stanford's main quad aerial view generated from ground-level shots), and bullet-time footage with just three iPhones.
- Robotics: Li argues robots' bottleneck is data, not chips—Atlas can port real environments into simulation with domain randomization.
- Roadmap: next steps are dynamics, editing and interaction; the team frames new-view prediction as a potential foundational primitive toward AGI.
Unlike the previous Marble model (bottlenecked by Gaussian splatting output), Atlas was designed multimodal from pretraining onward.
More from AGI Musings
- a16z: Superhuman AI hardware design is the easy part — manufacturing is the hard one — a16z · 2026-10-21
- a16z's Casado says AI can't do a VC's job — but it can do everything around it — alvelda · 2026-10-21
- Kent Beck: Let the Genie Carry the Toil, But Keep the Struggle — josh_wills · 2026-10-21
- Nearly $3B mapped: the money behind the AI existential risk ecosystem — KeyNaive2982 · 2026-10-21
- Three reasons AI is not conscious and may never be, per Nature essay — anilkseth · 2026-10-21
- Everything in AI is downstream of a global status graph, says founder — gabriel1 · 2026-10-21