Everything will be Simulated

Day 2 · Oct 14 Day 2 digest · More from this day

Runway co-founder Cristóbal Valenzuela argues that next-frame prediction, scaled far enough, turns video models into universal simulation engines, and demos Continuum, a video-model operating system that Runway plans to release within 8 to 12 months.

TL;DR

• Runway’s first AI film, “The Orange” (2019), came from a 128x128-pixel model force-upsampled to video. Valenzuela says today’s photoreal state of the art will look just as primitive in a few years.
• Scaled next-frame prediction learns physics, object permanence, gravity, and occlusion from video alone. Runway’s simulation-style environments have no coded physics laws — the model learns behavior “by brute force of scale”.
• Runway measures whether people can tell real video from generated video. Valenzuela says that Turing test for video is already passed, and benchmarks now weigh physics and geometry, not only taste.
• A “world interface model” turns any image into a clickable, interactive application with no Unity, Unreal, or Blender scene behind it.
• Runway is building a robotics lab so video models can drive autonomous, embodied systems.
• Continuum is an operating system whose screen is streamed by a video model: interactive “portals” cut from web content, a memory of past visualizations, and coding agents that reason visually. Most of it ships to everyone in 8 to 12 months.

Key lessons

1. Cinema’s 120-year-old trick is the core of the research. Video is an optical illusion: stitch frames fast enough and you get motion. Runway applies the same idea — next-frame prediction — with large networks and heavy compute, and rides scaling laws that it believes hold for video just as they do for language.
2. Emergent understanding comes free with scale. Predicting the next frame forces the model to learn physics, object permanence, gravity, occlusion, and how objects relate in a scene (his example: a camel standing on a ball, water in a container). Runway evaluates models on physics- and geometry-based benchmarks, not only on whether people like the output.
3. Real-versus-generated perception is measured and, per Valenzuela, already solved: “I look at that video, and I’ve been doing this for 10 years, I can’t tell you if that’s real or not” [08:33:52 | 5:14:27 PM]; the transcript renders the claim as “that touring test we already passed”.
4. Simulation emerges without physics laws. In game-like, playable environments generated by the model, Runway wrote no rules for object behavior — “The model just knows how to do that by brute force of scale” [08:35:54 | 5:16:29 PM]. Everything shown before the Continuum segment took minutes or seconds to render, so interactivity and real-time speed are the frontier.
5. Any pixel can become an interface. A “world interface model” — a video model trained to understand actions and clicks — made a single car image interactive: turn the lights, rotate the car, open the hood. There is no 3D representation: no Unity, no Unreal, no Blender. A second demo moved objects in a room from just an image.
6. Video models are becoming robot brains. Runway is spinning up an entire robotics lab; its New York office is half Hollywood studio, half robots walking around. The same world models are intended to power how autonomous, embodied systems walk and reason.
7. Continuum treats the operating system itself as a real-time video model. Selected web content becomes a “portal” — a fully interactive real-time video. Past portals form a memory (“the disk of the imaginations”). Resizing an application changes the experience. Searching for Mars made Continuum generate a Mars outpost interface in real time, seeded with data on robots currently on Mars and missions projected for 2030.
8. Agents can think visually, and extrapolation is a survival skill. Continuum connected to a coding agent lets the agent “think through” a problem visually (for example, a 3D model of a bike) instead of only via code, and lets humans guide the agent better than text alone. Valenzuela also argues that extrapolating the next 6, 12, 18 months is one of the most important skills in AI and company building today.

Watch this session

Cristóbal Valenzuela · 5:09:44 PM ET–5:26:46 PM ET · keynote · Watch on YouTube from 08:29:08

Notable quotes

• “I look at that video, and I’ve been doing this for 10 years, I can’t tell you if that’s real or not.” — [08:33:52 | 5:14:27 PM]
• “The model just knows how to do that by brute force of scale.” — [08:35:54 | 5:16:29 PM]
• “So when I say everything will be simulated, I’m really trying to understand or help not only our team but the world get to the point where a lot of what we see today will sound obvious, hopefully in a few more years or months.” — [08:45:35 | 5:26:11 PM]
• “So we’re really excited about is that for the next 8 to 12 months, we’re releasing most of what I showed here for everyone to use.” — [08:46:04 | 5:26:39 PM]

Session narrative

A stage voiceover opens the final keynote of the day: “What if AI could do more than generate content and actually simulate the physical world?” It welcomes Cristóbal Valenzuela, co-founder and co-CEO of Runway. He opens with a joke — “Welcome to my talk about the Matrix and how we’re living through it” — and grounds himself in place and time: almost a decade at Runway, started in New York about 10 years ago, building the company and its research lab there.

The argument starts with history. He screens Runway’s first AI film, “The Orange” (2019), built on an image model that generated 128x128 pixels and was forcefully upsampled. He calls it art now, “like looking at a film from the 1900s”. Then he cuts to state-of-the-art generated video: no real characters, no real locations. The point of the pairing is humility about the present — he still believes today’s output “will look like The Orange in just a couple more years”.

The core claim follows. Cinema discovered next-frame prediction 120 to 130 years ago; Runway scales that idea with big networks and heavy compute, backed by scaling laws the team believes apply to video. Predicting frames forces the model to learn physics, object permanence, gravity, occlusion, and object behavior. Runway measures whether humans can distinguish real from generated video and concludes that test is passed. Pushed into game-like environments, the model becomes a simulation system with no physics laws written by anyone — behavior learned “by brute force of scale”. He flags the frontier clearly: everything shown so far took minutes or seconds to generate, not real time.

The back half is a demo arc. First, a “world interface model”: one car image becomes an interactive application — lights on, car rotated, hood opened — with no Unity, Unreal, or Blender underneath. A second demo moves objects in a room from a single image. He also reveals a new robotics lab: half the New York office looks like a Hollywood studio, half is robots, with video models intended as the brains of autonomous, embodied systems. Then the closing extreme: Continuum, a research-stage operating system whose screen is streamed by a video model. He turns a fog-data website about San Francisco into an interactive “portal”, browses a memory of earlier portals (batteries), resizes a photosynthesis app to get a different experience, and searches for Mars, which makes Continuum generate a live, explorable Mars outpost seeded with real rover data. Last, “visual thinking”: a coding agent spins up Continuum to reason through problems visually, so a person can guide it better than through text alone. He closes by saying that when he says everything will be simulated, he means that this should feel obvious in a few years — and that Continuum may one day look like The Orange does now. Most of what he showed ships to everyone within 8 to 12 months.

Speaker details

Cristóbal Valenzuela - Co-Founder and Co-CEO, Runway. Cristóbal Valenzuela is the co-CEO and co-founder of Runway, a global AI research and technology company building foundational AI research models and tools to create real-world superintelligence. Before Runway, Cris was a researcher at NYU Tisch School of the Arts. Links: LinkedIn, Website

Full timestamped transcript

[08:29:08 | 5:09:44 PM] What if AI could do more than generate content and actually simulate the physical world?
[08:29:15 | 5:09:50 PM] To explore the next frontier of world models and what they could unlock, please welcome Cristóbal Valenzuela, co-founder and co-CEO of Runway.
[08:29:29 | 5:10:04 PM] Hey, guys.
[08:29:38 | 5:10:13 PM] Very, very excited to be here.
[08:29:42 | 5:10:17 PM] Thank you for staying for the last talk.
[08:29:45 | 5:10:21 PM] Welcome to my talk about the Matrix and how we're living through it.
[08:29:51 | 5:10:26 PM] So, you know, I've been working on Runway for almost a decade now.
[08:29:54 | 5:10:30 PM] Started here in New York, actually, maybe around 10 years ago.
[08:29:58 | 5:10:34 PM] And I want to walk you through some of the learnings we've seen through, you know, building the company.
[08:30:02 | 5:10:38 PM] You know, building the company, building this research lab here in New York, and where do we see a lot of the work that we do in video, in pixel generation, really moving forward?
[08:30:11 | 5:10:47 PM] And so, I don't kind of exaggerate and I joke that this might be a simulation or not.
[08:30:17 | 5:10:53 PM] And so, for maybe for perspective, I'll walk you through a piece of history.
[08:30:21 | 5:10:57 PM] This is like watching prehistoric ancient ruins.
[08:30:26 | 5:11:01 PM] This is our first AI film we ever made.
[08:30:28 | 5:11:03 PM] It was done in 2019.
[08:30:30 | 5:11:06 PM] It's called The Orange because, as you can tell very clearly, it's an orange that falls from the tree and walks through different parts of the world.
[08:30:39 | 5:11:15 PM] This is done—this is a video model, an image model that we had back then that was creating pixels in 128 pixels by 128 pixels that we then forcefully upsample to this kind of resolution.
[08:30:52 | 5:11:28 PM] I think of this as art now.
[08:30:55 | 5:11:30 PM] This is like looking at a film from the 1900s, you know?
[08:30:57 | 5:11:33 PM] It's beautiful.
[08:30:58 | 5:11:34 PM] There's a lot of history behind it.
[08:31:00 | 5:11:36 PM] But this is 2019.
[08:31:02 | 5:11:38 PM] We started the company in 2018, 2017, and so when we're trying to figure out if this had any legs, we were trying to convince ourselves that if you scale this, it will naturally improve to the point where we can get somewhere close to this.
[08:31:16 | 5:11:52 PM] This is state-of-the-art video today.
[08:31:18 | 5:11:53 PM] This is all generated.
[08:31:20 | 5:11:56 PM] None of these characters are real.
[08:31:22 | 5:11:57 PM] None of those locations are real.
[08:31:23 | 5:11:58 PM] I'm sure if you're probably here, you know this by now.
[08:31:26 | 5:12:02 PM] You probably have seen this online, on Twitter.
[08:31:29 | 5:12:05 PM] The models are just incredibly good.
[08:31:31 | 5:12:07 PM] And the most interesting thing is that when I look at stuff like this, I still believe this will look like The Orange in just a couple more years.
[08:31:39 | 5:12:15 PM] And so I want to walk you through, you know, how we're thinking not only through what we're doing at Runway today, but also in the next couple of years and the research that's shaping that.
[08:31:50 | 5:12:26 PM] We are a company that's at the intersection of both art and science.
[08:31:54 | 5:12:30 PM] We have deep roots in artistic and creative communities, and we have some of the best researchers in the world working on diffusion models and large video models.
[08:32:04 | 5:12:39 PM] And I like that combination because it allows us to, you know, bring the best of both worlds.
[08:32:09 | 5:12:45 PM] We're a team of scientists and artists thinking about the problem of scaling video and world models.
[08:32:15 | 5:12:50 PM] But in a way, the core idea of the company was already discovered by cinema and by moving pictures 120, 130 years ago, which is, if you think about video, and maybe you remember this from high school, video is an optical illusion.
[08:32:28 | 5:13:04 PM] Video is the illusion of movement, that if you do, if you stitch frames fast enough, you will create movement.
[08:32:35 | 5:13:11 PM] You will create the illusion of movement.
[08:32:36 | 5:13:12 PM] You create motion pictures.
[08:32:37 | 5:13:13 PM] And so at Runway, we're doing kind of something very similar.
[08:32:40 | 5:13:16 PM] We're taking that idea really to the next level.
[08:32:43 | 5:13:18 PM] How do we do next frame prediction fast enough, scale enough, in such a way that we can hopefully do more than just pretty and nice creative videos.
[08:32:54 | 5:13:29 PM] And so.
[08:32:55 | 5:13:30 PM] A lot is powered by, of course, scaling laws.
[08:32:59 | 5:13:34 PM] I think we believed in scaling laws not only apply for language models, but also for video, and this has been true.
[08:33:06 | 5:13:42 PM] We're now really ramping up the amount of compute that we put behind the models that you already have seen.
[08:33:12 | 5:13:48 PM] And I think the interesting thing about, you know, seeing the models scale over time is the emerging properties that appear, the intelligence that appears when you start taking this very simple idea that cinema found itself figuring out 120 years ago, which is next frame prediction, and taking next frame prediction really to the next level, right?
[08:33:31 | 5:14:07 PM] So if I take an image like this, which is not real, those two people are not real, and I just keep on predicting what the next frame will look like based on the previous frames, and if I do it fast enough, and if I do it with a network that's big enough, I'll get movement.
[08:33:46 | 5:14:22 PM] I'll get the illusion of movement.
[08:33:47 | 5:14:23 PM] I'll get something that looks so real to us that we humans can't distinguish.
[08:33:52 | 5:14:27 PM] I look at that video, and I've been doing this for 10 years, I can't tell you if that's real or not.
[08:33:58 | 5:14:33 PM] And we measure that.
[08:33:59 | 5:14:34 PM] We actually measure and try to quantify if people can distinguish between real and generated videos, and I think that touring test we already passed.
[08:34:06 | 5:14:42 PM] People are not able to distinguish.
[08:34:08 | 5:14:43 PM] But the most interesting thing is here, is that really what's happening as you predict next frames is you're learning a lot more than just pretty frames.
[08:34:16 | 5:14:51 PM] You're learning about physics.
[08:34:18 | 5:14:53 PM] You're learning about object permanence.
[08:34:19 | 5:14:55 PM] You're learning about gravity.
[08:34:20 | 5:14:56 PM] You're learning about occlusion.
[08:34:22 | 5:14:58 PM] You're learning about movements.
[08:34:23 | 5:14:59 PM] You're learning about how the objects work in the real world, and you're doing that all through just video, through images, through next frame prediction.
[08:34:33 | 5:15:08 PM] So if you look at a complex thing like this, a good video model, for it to understand how to make this a consistent video will need to naturally understand what all these objects are, how each of them relate to one another, how a camel might probably stand on a ball like that, what water in a container looks like, and if I do all of that together and I do it fast enough.
[08:34:56 | 5:15:31 PM] With fast enough, I'll get, hopefully, a consistent video that looks real to us.
[08:35:00 | 5:15:35 PM] So a lot of the benchmarks that we use to evaluate our video models are not only creative.
[08:35:07 | 5:15:43 PM] Creative basically means, do you like it or do you don't, are actually based on physics, on geometries, on how objects behave in the real world and how those objects behave in also simulator or digital worlds.
[08:35:21 | 5:15:56 PM] So you take that idea of next frame prediction and you do it fast enough, and you get, of course, now something that looks more like what we should expect from a video application or a game application.
[08:35:30 | 5:16:05 PM] Again, this is just video, right?
[08:35:33 | 5:16:08 PM] And you take that idea to the next level.
[08:35:35 | 5:16:11 PM] What happens if we start this idea of next frame prediction, but do it in the context of through the environment or something that looks similar to a game?
[08:35:42 | 5:16:17 PM] You can start actually getting closer to a physics engine, a simulation system.
[08:35:46 | 5:16:22 PM] But the simulation system has no physics laws.
[08:35:49 | 5:16:25 PM] We didn't describe what the behavior of the object should be.
[08:35:54 | 5:16:29 PM] The model just knows how to do that by brute force of scale.
[08:36:00 | 5:16:36 PM] And this kind of gets kind of interesting because you can only think about playing worlds.
[08:36:05 | 5:16:40 PM] This is not GTA.
[08:36:06 | 5:16:41 PM] This is someone playing someone else in an environment and a software that doesn't exist.
[08:36:13 | 5:16:48 PM] It's all simulation of a simulation, which gets really meta, right?
[08:36:17 | 5:16:53 PM] And so I'm not kidding when I said, welcome to my metric simulation.
[08:36:22 | 5:16:57 PM] But the most interesting thing for me is that everything you saw before was not real time, meaning that it took a couple minutes or seconds for those videos to be made.
[08:36:31 | 5:17:07 PM] A lot of what we're seeing right now, when we think about the next three, four years of video models and world models, is to deeply think about what happens when you start not only scaling the models and making them much, much, much bigger, with much more emerging properties, but also what happens when you start making them interactive, right?
[08:36:52 | 5:17:27 PM] So take the same idea of next frame prediction.
[08:36:54 | 5:17:29 PM] You take a frame, you predict the next frame, the next, the next, the next.
[08:36:57 | 5:17:32 PM] But then it can act at any point in time in that frame prediction, which means I can simulate any possible divergence or action in that world.
[08:37:05 | 5:17:41 PM] So in this world, I could play the video, but I can also come and touch every part, any aspect of that image and make it interactive to affect the consequences of that world.
[08:37:18 | 5:17:53 PM] And it works.
[08:37:22 | 5:17:57 PM] And of course, this is incredibly useful for a variety of different use cases.
[08:37:26 | 5:18:01 PM] As you can imagine, we're now spinning an entire robotics lab in Runway.
[08:37:31 | 5:18:07 PM] So if you come to our office in New York, half of it is what looks like a Hollywood studio, half of it is a bunch of robots walking through and moving through the office.
[08:37:39 | 5:18:15 PM] And so we're able to make these models, or start making these models, the brains of how autonomous and embodiment systems walk and reason about the world.
[08:37:48 | 5:18:24 PM] But there's a couple of more interesting use cases that I think go beyond the two obvious ones, which is video and robotics, which is what happens when you think about anything, any pixel on a screen, as a video model.
[08:38:02 | 5:18:37 PM] Here's a great example.
[08:38:03 | 5:18:38 PM] This is a video model.
[08:38:04 | 5:18:39 PM] What you see right now, well, this is an image, first of all.
[08:38:08 | 5:18:44 PM] That image could be real or not, but I could run one of our video models in real time and make this an interactive application.
[08:38:14 | 5:18:50 PM] So let's say I want to turn the lights of the car.
[08:38:18 | 5:18:53 PM] Maybe I want to turn the car around.
[08:38:20 | 5:18:56 PM] I can just click through it.
[08:38:22 | 5:18:57 PM] I can turn it around again, maybe open the hood and see what's inside the car.
[08:38:26 | 5:19:01 PM] Now, everything you just saw here has no physical 3D representation.
[08:38:32 | 5:19:07 PM] There's no Unity, there's no Unreal, there's no Blender.
[08:38:35 | 5:19:11 PM] It's just a video model.
[08:38:36 | 5:19:12 PM] A video model has been trained to understand actions and clicks in an interface.
[08:38:42 | 5:19:18 PM] We call this a world action model—sorry, a world interface model.
[08:38:47 | 5:19:23 PM] And the world interface model, well, I feel that you guys understand now, it's an interface and it's a world model—world interface model—allows you to do very cool stuff like this.
[08:38:57 | 5:19:32 PM] And so here's another application, another example of how you could think about using a model like this in real time.
[08:39:04 | 5:19:39 PM] I can take an image, a representation, move objects around, and expect them to move as if I was actually in that room.
[08:39:12 | 5:19:48 PM] Again.
[08:39:13 | 5:19:48 PM] Remember, there's no 3D representation of this.
[08:39:16 | 5:19:51 PM] This environment might not actually have been modeled before.
[08:39:19 | 5:19:54 PM] It's just an image and the model that understands how objects move in a space because of this very simple idea that next frame prediction scaled well enough can become a world model or a simulation engine.
[08:39:31 | 5:20:07 PM] So, with the eight minutes I have left, I want to show you this idea taken to another extreme.
[08:39:39 | 5:20:15 PM] We're thinking not only about what happens if you can control these cool applications and these cool interfaces, but what happens if you keep extrapolating.
[08:39:47 | 5:20:22 PM] And I think, by the way, extrapolating might be one of the most important skills these days in AI, in company building.
[08:39:55 | 5:20:30 PM] What happens over the next six months, 12 months, 18 months might be one of the most important things to try to predict and understand.
[08:40:02 | 5:20:37 PM] And so we asked ourselves that question a lot at Runway.
[08:40:05 | 5:20:40 PM] What happens?
[08:40:05 | 5:20:41 PM] What will happen if we keep scaling the models, if we keep adding more data, if we keep making them better simulation systems, if we make them universal simulation systems that can predict anything on a screen, literally anything on a screen?
[08:40:18 | 5:20:53 PM] So we've been working on this idea that we call Continuum.
[08:40:21 | 5:20:56 PM] And Continuum is the idea of real-time video models or world models taken really to the next degree.
[08:40:30 | 5:21:06 PM] This is an operating system.
[08:40:32 | 5:21:07 PM] The idea here is that you might open your computer at some point in the near future, and what you see streaming to your screen might be coming from a video model.
[08:40:38 | 5:21:14 PM] So I'll show you next specific how this idea could shape.
[08:40:42 | 5:21:18 PM] This is still very early research, but I want to show you this idea of combining video models as part of your operating system, as part of your daily work.
[08:40:53 | 5:21:28 PM] So here's an example.
[08:40:54 | 5:21:29 PM] I was researching, you know, San Francisco and the fog in San Francisco.
[08:40:59 | 5:21:34 PM] I wanted to learn more about when and how it happens.
[08:41:01 | 5:21:37 PM] This is structural data.
[08:41:02 | 5:21:38 PM] This is actually a website.
[08:41:04 | 5:21:39 PM] This is something you can take from the internet.
[08:41:08 | 5:21:43 PM] And now Continuum has this idea that I can select any part of content in that website and make it a portal.
[08:41:13 | 5:21:48 PM] We call this a portal.
[08:41:14 | 5:21:49 PM] And the portal is a real-time video.
[08:41:16 | 5:21:52 PM] It's fully interactive and fully based on the content you're experiencing, reading through.
[08:41:21 | 5:21:56 PM] I probably won't read through all of that text.
[08:41:23 | 5:21:58 PM] It's too much text for me.
[08:41:24 | 5:22:00 PM] So I just make a video, and I can make a video that's represented by the text you see there.
[08:41:29 | 5:22:05 PM] These portals can be 3D visualizations.
[08:41:30 | 5:22:06 PM] It can be any content that you're watching or interacting with.
[08:41:34 | 5:22:09 PM] On the right, you can have all the visualizations and portals you've experienced before, anything you've been trying to learn.
[08:41:40 | 5:22:15 PM] Basically, it's a memory.
[08:41:41 | 5:22:17 PM] It's the disk of the imaginations or the visualizations you've seen before.
[08:41:45 | 5:22:20 PM] So here I was researching about batteries, and so I can go back to the portal I had around batteries before.
[08:41:52 | 5:22:28 PM] And since they are real-time interactive applications, I can click them anywhere.
[08:41:56 | 5:22:31 PM] I can move them around.
[08:41:57 | 5:22:33 PM] I can start asking questions, and the videos and the content itself will change based on my interactions and the way I'm managing through it.
[08:42:10 | 5:22:45 PM] Here's another very cool example that when the first time I saw it, when the team showed it to me, it was like blew my mind.
[08:42:15 | 5:22:50 PM] Since it's a real-time video, you can make it fully interactive with anything, not only with clicks, but like resizing the application should also change the behavior, right?
[08:42:23 | 5:22:59 PM] So I can resize.
[08:42:24 | 5:23:00 PM] I'm learning about photosynthesis here.
[08:42:27 | 5:23:02 PM] If I resize the application, I get a different experience.
[08:42:31 | 5:23:06 PM] And since it's, again, full interactive applications, I can click through it, navigate through parts of the cell or how photosynthesis work to get a better understanding of what I'm trying to learn and do.
[08:42:42 | 5:23:17 PM] Again, the one thing that still feels to me kind of insane about this is I don't know how to call this.
[08:42:48 | 5:23:23 PM] I don't know if this is—it doesn't feel like a visualization, it doesn't feel like a video, it doesn't feel like a game.
[08:42:52 | 5:23:28 PM] It's something in between, and I think we need to do the collective work of figuring out how we call these kind of interactions in the first place.
[08:42:59 | 5:23:34 PM] And I'm sure you hear kind of the other thing is that you can change the structure of the prompt or what's actually creating that interaction in the first place.
[08:43:09 | 5:23:45 PM] Here is another example.
[08:43:13 | 5:23:49 PM] I'm going to type something, and this is one of the ones that I feel has been the most underexplored yet.
[08:43:23 | 5:23:58 PM] This is—we call this visual thinking.
[08:43:25 | 5:24:01 PM] So we've connected Continuum to a coding agent, and what the coding agent can do is it can use Continuum to visualize through a problem.
[08:43:33 | 5:24:08 PM] So imagine you ask your coding agent to create a 3D model for your bike, and instead of the model trying to just think through that via code, the model can spin a version of Continuum to think through the problem visually.
[08:43:45 | 5:24:21 PM] So if you ever think about someone—I'm sure you have friends or maybe a lot of people here—maybe visual thinkers.
[08:43:51 | 5:24:26 PM] What a visual thinker is, is that— Think through problems and through ideas visually first.
[08:43:56 | 5:24:32 PM] And that helps you ground and project some of the best things you want to do.
[08:44:02 | 5:24:37 PM] And this is somehow very similar.
[08:44:04 | 5:24:40 PM] Agents have the ability to do visual thinking.
[08:44:07 | 5:24:43 PM] Agents can spin up a sort of imagination that you can tap and interact with.
[08:44:12 | 5:24:48 PM] So not only the agent can interact with it, I can also see it myself.
[08:44:16 | 5:24:52 PM] And I can guide the agent in much better ways than I might have been able to guide it just via text alone, which is for us a very powerful idea.
[08:44:25 | 5:25:00 PM] It's a completely new paradigm.
[08:44:29 | 5:25:04 PM] We call this idea visual thinking.
[08:44:33 | 5:25:08 PM] The last example I have is how you think about taking, you know, Continuum to become much more than just using or navigating interactive structural content in the way I've showed you before, but actually making them fully live experiences that are reactive.
[08:44:51 | 5:25:27 PM] So in this case, I was searching for Mars, and as Continuum understands the intention of what I'm trying to search from, it also suggests interfaces that are specifically designed in real time just for that.
[08:45:05 | 5:25:41 PM] So here it created for me this Mars outpost that actually should be accurate with the data about the actual robots that are in Mars right now, or the ones that are projected maybe in 2030.
[08:45:16 | 5:25:52 PM] And since this is fully interactive, I can interact with it.
[08:45:19 | 5:25:54 PM] I can go to places.
[08:45:20 | 5:25:55 PM] I can move around in the same way that I might move around in a 3D system or in a 3D continuum.
[08:45:28 | 5:26:03 PM] And it works.
[08:45:30 | 5:26:05 PM] Just fascinating.
[08:45:35 | 5:26:11 PM] So when I say everything will be simulated, I'm really trying to understand or help not only our team but the world get to the point where a lot of what we see today will sound obvious, hopefully in a few more years or months.
[08:45:49 | 5:26:24 PM] We look back at Continuum and it might feel very close to the video of the OR that we just showed you.
[08:45:54 | 5:26:29 PM] Continuum and a lot of the things we're doing in video and in runway when it comes to simulation are still in very early days, but we're getting to a point where it actually works and it really scales.
[08:46:04 | 5:26:39 PM] So we're really excited about is that for the next 8 to 12 months, we're releasing most of what I showed here for everyone to use.
[08:46:10 | 5:26:46 PM] Thank you, guys.