Justin Johnson

We are excited to announce that World Labs is joining @AMD.

The research and technical breakthroughs we have achieved since our founding in 2024 have given us a clear vision for AI’s potential to solve problems in the spatial and physical world.

Accelerating the future of spatial and physical intelligence requires scaling our efforts, scaling our reach, and getting closer to the hardware.

One of my favorites: Sparse reconstruction powered by Atlas.
The full scene was reconstructed from only 19 photos I took with my phone.

#ECCV2026 Test of Time Award winners

One more thing…

Always start your day right with a fresh bowl of Atlas.

Atlas > Fable

Biggest AI breakthrough in a long time.

ok this one was challenging! i guess SEM has an orthographic camera model, ended up getting posed as a very narrow FOV perspective cam and having some alignment errors as a result.

if there was calibration data available for the ray bundles it could potentially help (though the model was definitely never trained on data like this!)

I'll be on MTS later today, excited to talk about the latest from World Labs!

one of my favorite things about Atlas is its ability to blend reconstructing the real-world with generating imaginative worlds. the possibilities are endless.

so proud of our team, can't wait to have everyone try it out in the next generation of @theworldlabs 's marble :)

In this episode, @jcjohnss, co-founder of @theworldlabs, joins us to discuss world models and the emerging field of spatial AI. We explore why many researchers see capabilities beyond language as an important frontier for AI, and what it means to build models that can understand, generate, and simulate the environments around them.

Justin explains the different approaches to world modeling, including explicit 3D representations and generative models, and why there is still no established recipe for building these systems. We also discuss World Labs’ Marble system, which can generate navigable 3D worlds from images and other inputs, the challenges of evaluating world models, and the role of simulation, planning, and action. Finally, Justin shares his vision for models that bring these capabilities together, supporting everything from interactive virtual environments to agents and robots that can operate in the physical world.

🗒️ Full show notes: twimlai.com/go/775.

📖 CHAPTERS
===============================
00:00 - Introduction
03:19 - Defining World Models
08:10 - World Models as Theory Builders
11:37 - POMDPs and Agent–World Interaction
15:40 - Training Agents with Behavior Cloning
18:40 - Ground-Truth State and Learned State
23:34 - Explicit 3D vs. Implicit 3D
28:14 - Reconstruction vs. Generative World Modeling
30:38 - Gaussian Splat Anatomy and File Formats
33:52 - Why Gaussian Splats Work with Neural Networks
37:00 - Consistency by Construction and at Scale
40:31 - How Marble Generates 3D Worlds
44:25 - Training Data and Output Representations
47:48 - World Models as Renderers, Planners, and Simulators
51:30 - When Rendering and Simulation Overlap
56:28 - The Path to Unified World Models
59:30 - Architectures, Loss Functions, and Long Contexts
01:02:47 - Where to Learn More

I’m so excited to finally talk about our new model, Atlas, and share some generations from it..so excited, in fact, that I made a fan trailer for it 😆

It’s an incredibly versatile, multi-task model with a wide range of capabilities. There are many powerful video models out there right now, but after working with Atlas and experimenting with it for so long, I’ve come to really appreciate its elegance and malleability...all rooted in the framework’s simple autoregressive design.

As you can see in this video, Atlas is, at a high level, a camera-conditioned visual generation model. But its formulation allows it to learn and perform a variety of tasks remarkably efficiently!

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.

Massive stream today on MTS, starting with a review of the news dominating the timeline:

- Physical Superintelligence comes out of stealth w/ a $58m seed round
- Bernie Sanders' Fox op-ed calling for an international pause on AI
- OpenAI's support of California SB 1119
- SB Energy's IPO filing

Then straight into some amazing interviews throughout today:

- @jcjohnss co-founder @theworldlabs to discuss how world labs is evolving world models & what to expect in the future

- @TaylorLorenz tech journalist @usermagazine to talk about the anthropomorphizing of AI and why this has been such a divisive topic

- @hosseeb managing partner @dragonflyvc to talk about the rogue agent discourse and crypto's role in agentic self preservation

- @tsmilton_amodo ceo at amodo design to talk more about building the 21st century's most important hardware, and the role of ai in engineering/development

- @adamkovac founder and ceo @progresschamber to talk about the similarities between data center opposition and fracking

- @ruxandrabio writer @WorksInProgMag to talk about her most recent article on why even curing cancer might not redeem AI

- Lance Co Ting Keh, co-founder and ceo of @t0finance on how he's working to transform accounting and financial intelligence for all startups

- @aditabrm co-founder and ceo of @reductoai to talk more about their newest doc parsing model r-1 & why now is such an exciting time to be working with unstructured data/documents

- @JeffImmelt venture partner @NEA to talk about his work investing and how ai is transforming so many startups & existing business

- Benjamin Jaros, research fellow @HooverInst to discuss some of his writing on the billionaire tax & why this poor policy proposal has gotten so far in American political dialogue

The world is not just made of words, and spatial intelligence was never just about perceiving and generating worlds. It's about interacting with them.

Today, SceniX is joining World Labs. 🌎🤖👇

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]

1/ Introducing GPIC: a Giant Permissive Image Corpus and benchmark for visual generation!

🚀100M VLM-captioned image-text pairs for training
📊1M image-text pairs for benchmarking
🖼️~28 trillion pixels
🤗Centrally Hosted
✅Fully permissive for research + commercial use

Dataset, benchmark and models🧵👇

Co-led with @KyleSargentAI

What's the right space to diffuse in: Raw Data or Latents?

Why not both!

In Latent Forcing, we order a joint diffusion trajectory to reveal Latents before Pixels, leading to improved convergence while being lossless at encoding and end-to-end at inference.

w/ @drfeifei+...
1/n

The World API is live.

Generate persistent, explorable 3D worlds from text, images, and video.

Integrate them directly into your products.

Introducing Marble by World Labs: a foundation for a spatially intelligent future.

Create your world at marble.worldlabs.ai