Skip to main content
SpAItial
Back to Blog

Introducing SpAItial: AI that creates and understands 3D worlds

We're launching SpAItial: a new AI company building systems that create and understand 3D worlds. Learn about our Spatial Foundation Models, AI models that work directly in 3D space.

SpAItial founders

Generative AI is reaching the point where machines can do more than understand pixels or words. Today, we're introducing SpAItial: a company built to take on one of the hardest and most important problems in AI, which is building intelligent systems that natively understand and generate 3D worlds. With an experienced team and leading investors behind us, we're setting out to change how AI understands space, time, and the physical world.

SpAItial is building Spatial Foundation Models (SFMs), a new kind of AI model that generates and reasons about the appearance and physics of real and imagined environments. SFMs have a built-in understanding of space and time, which makes new applications possible where the virtual and physical worlds meet.

Unlike existing generative AI, such as large language models, image models, or video models, SFMs work directly in physical space. Working in 3D makes them much better at reasoning about the world and closer to how people understand it. SFMs can change industries ranging from immersive entertainment and computer-aided design (CAD) to VR/AR and robots that understand the physical world.

Interactive 3D has clear benefits for any kind of visualization. Rendered at photorealistic quality, it is more engaging, immersive, and faithful than 2D content.

Our mission: Connecting the virtual and physical worlds

At SpAItial, we envision a future where AI understands and connects the virtual and physical worlds. Our mission is to build AI that follows the laws of physics, is grounded in space and time, and understands the 3D world the way people do. Our SFMs work directly on 3D structures and model geometry, materials, and physics from the start. This lets us generate photorealistic, physically plausible environments from sparse inputs such as a single image, a short video, or even a text description.

We're designing SFMs for uses that range from building high-fidelity game environments to helping robots navigate spaces that change:

  • Immersive media and entertainment: Gaming, film, and AR/VR experiences with coherent, photorealistic 3D environments.
  • Digital twins and design simulation: Real-time, high-fidelity replicas of physical environments for planning, visualization, and testing.
  • Autonomous systems and robotics: Better spatial reasoning, navigation, and interaction in physical spaces.
  • Industrial and urban infrastructure: Smarter modeling, automation, and control across construction, logistics, and manufacturing.

Our team

Our founders combine academic research with experience building and scaling real products:

  • Matthias Niessner: Professor at TU Munich, co-founder of Synthesia, formerly at Stanford.
  • Luke Rogers: Formerly at McKinsey and an early leader at Cazoo, where he helped scale the company from seed to IPO.
  • Ricardo Martin-Brualla: Formerly at Google, where he most recently led generative 3D work at Google Shopping and, before that, ML at Google Beam.
  • David Novotny: Formerly at Meta, where he led work on 3D asset generation and digital twinning.

Beyond the founders, our wider technical team has deep expertise in AI research, large foundation models, graphics, and 3D generative AI. It includes Katja Schwarz, formerly at Meta, who developed early generative 3D models like GRAF and VoxGRAF, and Mojtaba Bemana, whose X-Fields paper defined 4D neural representations and earned the 2025 EuroGraphics PhD Award. We've built a strong team that shares one goal, and we're just getting started.

SpAItial team

Our seed round

We've raised a $13 million seed round led by Earlybird Venture Capital, with Speedinvest and a group of prominent angel investors also taking part. We'll use the funding to develop our models faster, scale our infrastructure, and grow our team.

The angel investors include Robin Rombach (Black Forest Labs), Victor Riparbelli and Steffen Tjerrild (Synthesia), Christian Stiebner, Edward Grefenstette, Maximilian Odendahl (ex-CEO Silexica), Jon Barron, Rahul Garg, Sergej Epp (CISO at Sysdig and Palo Alto Networks), Elias Schneider (Codesphere), Paul Whitehead (Zoopla), and others.

We see their support as a sign that we're on the right track, and we're eager to keep building.

What's next: Scaling and collaboration

We're focused on improving our models, building the technology we need to deploy them at scale, and hiring more people in AI, graphics, and 3D systems. At the same time, we're partnering with leading organizations to test our technology on real-world applications across industries.

Echo is now available in the SpAItial app. If you're interested in trying our models or collaborating with us, .

We're hiring

We're just beginning to change how AI sees and interacts with the world. If you want to work on spatial intelligence, we'd love to hear from you.

View open roles

A sneak peek

We show our model generating 3D environments from simple text prompts or images. You can render these environments interactively from new camera viewpoints, at photorealistic quality. This teaser is a first look at what our early model can do and where we're heading.