Introducing Atlas:
The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D.
Model the world, move the camera, and simulate space & time.
After 2.5 remarkable years building World Labs, I have decided to step away from my role and will continue supporting the company as an advisor.
The last 2.5 years have been some of the most intense and meaningful of my life. Building alongside such an exceptional team has been one of the great privileges of my career. Together, we helped bring the idea of spatial intelligence closer to reality: building the systems, models, and foundations for a new frontier in AI.
Over the past few months, several personal events, including an accident that resulted in multiple fractures and a concussion I am still recovering from, gave me the time and perspective to reflect. Seeing the company in such a strong position, with Marble helping demonstrate where world models can go, gave me confidence that this was the right time to step back, prioritize a full recovery, and create space for what comes next.
This was an incredibly difficult decision, first and foremost because I believe deeply in World Labs and the importance of what we are building. I’m incredibly grateful to Fei-Fei, Justin, Ben, and the entire team for the opportunity to build together. I’m also deeply thankful to the partners, creators, and collaborators who helped bring this vision to life and have been part of this journey with us. World Labs and this team will always hold a very special place in my heart, and I’ll be cheering for them every step of the way.
I don’t yet know exactly what the next chapter holds for me, but I’m excited for what’s ahead. I’ll share more as that journey unfolds.
This is the note I shared with the World Labs team yesterday:
Today we are sharing three new research papers, each exploring a new way to generate 3D content by leveraging large-scale generative models and 2D priors.
These projects were led by our incredible interns @HaoZhang623 @BDuisterhof @DrTunnels
[1/4]
The Ring (Montreal) — Dormant Memories series
New alternate scene, and an iteration on the UX and flow.
Starting from a volumetric capture of a real location in Montreal, reconstructed as 3D Gaussian Splats, I then used @theworldlabs' Marble to generate alternate versions of that same space.
Running in real time using @sparkjsdev.
#GaussianSplatting #3DGS #pointcloud #sparkjs #threejs
Text became the universal interface for software.
3D is becoming the universal interface for space.
It’s the medium that lets humans and AI generate, edit, simulate, and share worlds together.
https://t.co/hd0PMNTv6j
Spark 2.0 Developer Preview is now available: introducing Level-of-Detail (LoD) rendering and streaming for world-scale Gaussian Splats on the web.
Learn more ↓
World Labs has raised $1 billion in new funding. We are grateful and excited to partner with our investors, including AMD, Autodesk, Emerson Collective, Fidelity Management & Research Company, NVIDIA, and Sea, among others.
https://t.co/MfK7mLGKPH
From a single image to a full robotics simulation.
360 image → Marble → 3D environment → Robotics sim
This is how we scale world creation for embodied AI with @LightwheelAI, turning lightweight capture into simulation-ready environments in minutes, not weeks.
Researchers are exploring Marble’s generative 3D worlds as a way to rapidly produce simulation-ready environments for robotics without manual scene construction.
World Labs and VIVE Mars are making it easier to turn ideas into shoot-ready 3D scenes, fast.
@theworldlabs new AI world-builder lets you create detailed virtual environments from a simple text or image prompt, then composite them live. Once the scene is ready, you can step straight into production using VIVE Mars CamTrack.
Sign up at a discounted rate to start building worlds, and see how the full workflow comes together with VIVE Mars CamTrack and Marble.
👉 https://t.co/TPwCMrE938
👉 https://t.co/q18B8CrRhW