Matt Wolfe reacts to a demo that generates full 3D scenes from text prompts. The scenes have real 3D geometry you can explore and open in Meta Quest VR. This could soon let filmmakers and game developers build entire worlds from a single sentence.
This is just still blowing my mind.
Like, this is so wild.
A single prompt generated this 3D room that I can move around in. I can look behind things that weren't in the original image. We're now moving from 2D images and videos into full 3D world generation by AI Marble's.
Allowing us to generate.
Explorable 3D scenes from text prompts. And it understands the 3D geometry.
Of the scene.
And if you have a meta quest, you can actually open this in VR. I've experienced it. It. It's crazy. And then if you want, you can actually take a bunch of these worlds and string them together in a much larger world. This is one that you got to jump into, play.
With it for free and see what it's capable of. Because this is wild.
This is probably the future of asset creation. I imagine full game worlds and background environments for films are going to be prompted into existence in the near future. I mean, we can do it now, but this is only going to get better.
AI News: Dots, GPT-6.1 Sol, Sonnet 5.5, Gemini 4, and everything you need to know
