3D ($DDD) podcast mentions
I guess I'm thinking if like, do you have either benchmarks as you're fine-tuning models, testing models, and do you notice anything particular about models where you're like, oh, it's spiked on this particular vector of fraud detection, do you notice the spiky intelligence behaving differently? Because the X algorithm will be obsessed with a particular model typically by how viral the creation can be. So if it's a Blender 3D model or a video game, everyone crowns that the best model of the day. But no one's really like, wow, this model is incredible at detecting financial fraud because it's probably just collapsed into some sort of score. But what are you doing on actual model benchmarking evaluation? And then how do you communicate that to your customers?
John Coogan · TBPN · Third Party Evaluators, Fed Hikes Rates, Zuck Pushes Back On AI Slowdown | Jeremy Allaire, Tomasz Tunguz, William Layden, Justin Beroz, Eli Wachs, Sean McCarthy, Tom Mueller · 2026-09-16Yeah, absolutely. So this is great work on behalf of the geologic team headed up by Eric, Eric Corbin, the two Bens who do our 3D modeling, Zach, who's helped us out, Henry as well. That whole team has done an excellent job over the past, call it 18 months to 24 months. Of targeting and putting all this 3D modeling together, a 3D geochemical model, MT survey, uh, mag, did the geophysics, got us to this point where we're now understanding this system better than anyone ever has before. I mean, we came into this and said it was going to be the most well-informed drill program. Uh, and certainly as we're on the hole 2, uh, I think they're doing an excellent job of proving that right. There is no drilling off to the southwest. Uh, it is the first time that we've shown that the structures are heading that way. And so we'll have a follow-up hole on that. Uh, we'll probably start drilling that in roughly 2 weeks' time. Uh, we're just finishing up some other holes that are down south on the project, uh, and then we'll move the drill rig back up towards where we drilled previously. So incredibly excited and, uh, excited to see what drilling brings us out that way.
Warwick Smith · Mining Stock Daily · American Pacific Mining's High-Grade Step-Out Points to a Larger System at Madison · 2026-09-14As long as, well, set aside sound, whatever, but you basically have the full power of a computer there. And we had some aborted attempts early on to try to build agents that could do that. And so it really took us until now, but you're seeing the power immediately. And it's just been so cool to see people take the Blender capabilities and, you know, take a screenshot of something and then have it make a 3D model and you can actually, you know, lots of people are now designing houses or trying to redesign, uh, their living room, all those things by just utilizing this capability. And to me, the, the, the thing that really stands out is that you can now move forward on AI that can do things for you without you having to build all this specific connectors. And I think there's so much software they don't even think about that you have to orchestrate every day. And like, how much of your life is like clicking around menus and like, You know, typing things into a spreadsheet and things like that. Like none of that is what we should be doing. 100 years ago, no one is doing any of these things. And so it's not crazy to think that in 5 years, 10 years, no one will be doing any of this stuff anymore. That we will get our time back. We're not going to have to be getting our carpal tunnel or hunched shoulders or, you know, all of those, all of those physical problems that are us contorting to the machine. It's now the machine is there to help us, to empower us, to, to really serve us.
Greg Brockman · The a16z Show · Greg Brockman on Why OpenAI Says We’re Entering the AGI Era · 2026-09-14We are live with Justin Johnson, who is a co-founder of World Labs, which is a startup that builds spatial intelligence products. And just today they announced 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 and time. So congrats on the launch. Like, the launch video looks really cool. Welcome to MTS.
Theo Jaffe · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Right. So there, what Atlas can do is sparse reconstruction. I can take just a few photos of a space, as few as one, up to 100 or more. And the more photos you give it, the more you can reconstruct that existing real-world space accurately. And the third thing we can do is simulation, right? And this is going after more robotics. Yeah. Well, both VFX and robotics, but there we want to be able to take an existing space, recreate that space in 3D, then simulate what a robot might— how a robot might behave in that space. And we show some examples of all these different use cases on the blog.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, that's a great question. So here we built Atlas specifically around spatial control and the notion of a spatial context. So when we get, when you give Atlas reference images, they aren't just images. They're actually grounded in 3D space. So you can stage those images in 3D and say, this reference image should be here, this reference image should be here, this reference image should be here. Then when you generate a video from Atlas, you're not just gener— you're not just giving it a text prompt. You also go in and steer the camera and say, I want the camera to fly this way, this way, this way, this way with exact precise pixel control. So then we show some examples where even up to a minute of video, um, we give it more reference images and sort of leave breadcrumbs of those images along the trail in 3D. And then you get that directorial control over the video that you're generating. So the idea here is that even as you go to really long generations, um, it's not a slot machine. You're generating and directing and understand— really controlling what the model outputs.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, I think so. I think there's actually two different futures here of like the tech path that gets us there. One is sort of the explicit 3D route that you mentioned, right? Like the idea of generating a Gaussian Splat reconstruction, then importing that explicit asset into a game engine or something else. I mean, this is something Atlas can do, right? Like it can be used to generate, you know, 3D Gaussian Splat scenes and then you could import them and generate and build games around them. Another direction where this could go is, you know, you skip the engine entirely and go directly for frame generation, right? And I have the model directly outputting the frames without any explicit 3D in the mix anywhere. Um, and Atlas is, is built to do this as well. So it's, it's one model that both handles, you know, 2D when you need it, 3D when you need it. Um, and it can handle both modalities in a native way.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, we have some internal tooling for this, but it, it's gonna take a little bit more, uh, little, little bit more iteration on the product side to figure out the right way to expose that. Um, but the internal tooling is, is actually pretty awesome, right? You can basically take images, drag, drag them into this interface, have this sort of spatial 3D canvas for all your input images. Um, then they end up with this like proxy geometry in 3D space, and you can steer your camera kind of with an FPS type control. To decide exactly where you want this thing to go. And that's a pretty complicated user experience or user interface. So I think we'll probably iterate on different ways to expose these capabilities in nice ways.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, so it's a totally rebuilt stack of models, right? I think the central thesis is the same. We want to build multimodal models that can input many different modalities, output many modalities. This is something that Marble did as well. But one way in which they differ is that Marble really put Gaussian Splats really at the center of many things that it did. So Marble could output videos, Marble could output 3D scenes, Marble could output meshes, but everything was kind of bottlenecking through a 3D Gaussian Splat representation. So whenever we output videos or images from Marble, you were always generating a Gaussian scene, then rendering those Gaussians back to images or videos. Um, and this is really nice, but, um, we wanted to break that dependency. So Atlas, you know, was built fully from the ground up to both rethink the architecture to make it more unified, more easy to scale, um, and to have it have that sort of that branch between 2D and 3D happen earlier, right? So that when we're generating 2D outputs or video outputs from Atlas, you're not bottlenecking through a Gaussian Splat scene anymore. The model is directly giving you those beautiful 2D pixels. And then only you lift them up to 3D only for the applications that you, that you actually know that you need 3D.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13I think Gaussians are a tool just like anything else, and you got to use the right tool for the job, right? Like if you're working on something like VFX, then the ultimate output is a 2D image and, or a 2D image or 2D video, and maybe that's all you need. But for other applications, especially real-time applications on embedded devices, like I think Gaussians and other explicit 3D representations still have their place, right? I think there might be a world in the future where we're all streaming pixels directly off these models in real time from powerful server farms, but You know, that's— we're not quite there yet. And even if we are, it's going to be pretty expensive. So something like a Gaussian or a mesh is really useful for client-side rendering, especially on mobile or VR devices, and also really useful for integrating with existing workflows and pipelines, right? Like it's, you know, I think we all believe in the future of technology and we're excited about building these solutions that, you know, give us whole new experiences. But the reality is that it takes a lot longer than these— it takes a lot longer for these things to permeate than we expect. right? So people in gaming and VFX and design and architecture, they have their existing workflows that work on 3D representations. And, you know, instead of trying to switch them over all at once to a fully AI-native thing, if you can get with AI into their workflows and meet them where they are, then that's another great use of, of Splats and other 3D, uh, and other explicit 3D representations.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Got it. Yeah. One of my favorite video games ever is Microsoft Flight Simulator. Yeah. Which is you get in a plane and you fly around the world and it, uses like a lot of, not quite real time, but it uses satellite data in order to like reconstruct a 3D model of the world basically in reasonably high fidelity, not incredibly high, reasonably high. Um, do you think Atlas will eventually be able to, uh, like, like what will it take to get Atlas from where it is now to able to dynamically generate content quickly enough to, uh, do what Flight Simulator does? And like, is, is this even like the right tech tree for that?
Theo Jaffe · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13I mean, I think it'll happen sooner than you expect, right? And that'll happen in one of two ways. Like there's one sort of, and actually, you know, Atlas was built to handle both of those tech trees. One is the explicit 3D route, right? Like maybe we're generating a Gaussian scene or a 3D point cloud scene and then interpolating that on the client. Um, that's something Atlas can do because of its native 3D outputs. The other version, like I said, is just directly streaming the frames, right? And that's something that I think we'll be able to do pretty soon.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, definitely. And that's, that's what we want to do. So we want to get this to, that's, that's kind of like suits, fits into our thesis about these, these models as horizontal general things. So if we want to expose these models to a lot of people, let them be consumable by humans and agents alike, and let them interoperate with other tools. So if we can provide a model that has all these general capabilities in simulation and reconstruction and generation in explicit 3D and 2D, and then we can have a coding agent go and stitch all these capabilities together, then we can let people really quickly iterate on really cool experiences. So I'm really excited to see, once we get this into people's hands, what kinds of cool experiences people will build with it.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, I can imagine like a Pixar but in 3D, like, uh, in like a 3D navigable space. Sort of thing emerging. Like, I think there's— this is going to be really good for creatives specifically.
Sophia Puccini · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13Yeah, I think so. Right. And what we've wanted to build on all these things is this aspect of creative control, right? We never wanted to be, to be building these generative slot machines where you just like pull the thing and hope you get out a good generation. Um, or, or you don't, you don't feel like that generation was yours or that you controlled it or that you knew what you were getting out. Like, we want to build these tools that have really deep control, right? Like whether it's 3D control, whether it's text control, that you feel like you're a director in the thing guiding this model and deciding exactly what it outputs rather than just like the model spits something out and maybe I can stitch it together after the fact.
Justin Johnson · The a16z Show · World Models, Robotics, and the Future of 3D AI · 2026-09-13