# Robots Will Train Mostly in Simulation by 2027 | Neil Nie

Markdown copy of https://realalpha.app/v/zONxIHfiY9k/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

Neil Nie predicts robots will shift toward simulation training within a year and says local models will handle fast, safety-critical control.

- Show: MTS
- Hosts: [Sophia Dew](https://realalpha.app/people/sophia-dew/)
- Guest: [Neil Nie](https://realalpha.app/builders/neil-nie/) is CEO of Verne Robotics.
- Published: Nov 15, 2026
- Length: 22 min

## The best minutes

- [3:31](https://realalpha.app/v/zONxIHfiY9k/?t=211) [Robotics can inherit useful knowledge from non-robotics data](https://realalpha.app/s/robotics-can-inherit-useful-knowledge-from-non-robotics-data/) (4 min)
- [8:14](https://realalpha.app/v/zONxIHfiY9k/?t=494) [Robots need local models for fast, reliable control](https://realalpha.app/s/robots-need-local-models-for-fast-reliable-control/) (6 min)

## Also worth watching

- [15:11](https://realalpha.app/v/zONxIHfiY9k/?t=911) [V0.5 combines full-body control with task-state awareness](https://realalpha.app/s/v0-5-combines-full-body-control-with-task-state-awareness/) (2 min)

Skip the 3-minute company and open-weight discussion, the model-leadership exchange, the product update, and the closing sponsor reads.

## Full analysis: Neil Nie expects more robots to train in simulation within one year

- Nie expects a lot of robots to do more training in simulation within one year.
- That is a forecast without a baseline, adoption share, or named task.
- Teams should measure simulated training against physical validation and deployed results.

Its 7 sections, what it means for you and what to watch are on [the page](https://realalpha.app/v/zONxIHfiY9k/#analysis), free with an account:

- [17:54](https://realalpha.app/v/zONxIHfiY9k/?t=1074) Nie forecasts more simulated training, without a share
- [18:54](https://realalpha.app/v/zONxIHfiY9k/?t=1134) His commercial exposure calls for evidence, not dismissal
- [18:00](https://realalpha.app/v/zONxIHfiY9k/?t=1080) Simulation needs physical checks before it informs deployment
- [17:58](https://realalpha.app/v/zONxIHfiY9k/?t=1078) Other data problems frame, but do not test, the forecast
- [10:17](https://realalpha.app/v/zONxIHfiY9k/?t=617) Local policies add a separate deployment constraint
- [12:48](https://realalpha.app/v/zONxIHfiY9k/?t=768) General models may absorb specialized robotics advantages
- [17:10](https://realalpha.app/v/zONxIHfiY9k/?t=1030) Verne's model features do not prove the forecast

## Summary

Neil Nie is CEO of Verne Robotics, which released V0.5, an open-weight world action model for industrial robots. He says the company is already deploying robots with customers and is expanding its team.

Nie argues that robotics will rely on both fast local models and larger cloud models, with local inference helping robots respond despite latency or Wi-Fi problems. He says GPT-6 Astra’s ability to control robots with little robotics-specific training marked a watershed moment, while specialized models still have advantages on speed and cycle time. He supports open-weight models as a way to broaden deployment and sees V0.5 as suited to full-body tasks and agentic systems. Nie predicts that within a year more robots will train in simulation than on real-world data.

### Minute by minute

- [0:50](https://realalpha.app/v/zONxIHfiY9k/?t=50) **Open weights can accelerate deployment.** Nie says Verne’s capture device records customer tasks such as grasping packages and assembly for model training. He describes V0.5 as one of the leading U.S.-made open-source foundation models and argues open weights can lower adoption barriers, support U.S. leadership and serve national-security interests.
- [3:31](https://realalpha.app/v/zONxIHfiY9k/?t=211) **Astra changed robotics expectations.** Nie agrees that GPT-6 Astra was a GPT-3 moment for robotics. He says its ability to control a robot without much robotics-specific training showed that knowledge from 3D control and video games can transfer to robotics. He expects fast reactive models and more capable models to work together.
- [6:11](https://realalpha.app/v/zONxIHfiY9k/?t=371) **Deployment relationships create the moat.** Nie says open models let customers inspect and benchmark systems, while much of a robotics company’s competitive moat comes from customer relationships and deployment. He sees integrators using open models to help labor-constrained, mid-sized factories and warehouses increase output as well as reduce costs.
- [8:14](https://realalpha.app/v/zONxIHfiY9k/?t=494) **Safety favors local inference.** Nie says fast robot-control models should run on-device to avoid latency and Wi-Fi disruptions, while larger models can run in the cloud at a slower pace. He describes humanoids processing camera images on a GPU with a model of almost 10 billion parameters and producing actions five times per second; he says about half the companies he saw at a conference ran policies locally.
- [12:00](https://realalpha.app/v/zONxIHfiY9k/?t=720) **Robots will combine local and cloud models.** Nie names NVIDIA Groot and Physical Intelligence’s model as strong local options. He says GPT-6 Astra does better in unseen environments, while specialized robotics models can win on cycle time and speed; he expects general models to absorb more tasks and says NVIDIA remains the category leader in local open-weight models.
- [15:11](https://realalpha.app/v/zONxIHfiY9k/?t=911) **Video model teams are entering robotics.** Nie says solving robotics problems draws on multiple areas of machine learning, but remains robotics-driven. He points to Runway and Black Forest Labs entering the field and says competition can improve open-weight models. He describes V0.5 as a full-body control model designed for agentic systems that can track task completion.
- [17:21](https://realalpha.app/v/zONxIHfiY9k/?t=1041) **Simulation could overtake real data.** Nie predicts that within a year many robots will be trained more in simulation than through real-world data collection. He expects a strong real-to-sim-to-real flywheel and says simulation is an area Verne is betting on.
- [18:01](https://realalpha.app/v/zONxIHfiY9k/?t=1081) **Verne expects deployment growth.** Nie says Verne plans to launch more products and expand its team across engineering and deployment. He calls 2026 a year for models and research and expects 2027 to be a year of deployment. The company is already deploying with customers and aims to triple that in the coming quarters.

## Transcript

MTS, hosted by Sophia Dew, with Neil Nie.

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[0:00](https://realalpha.app/v/zONxIHfiY9k/?t=0) simulation will become more and more relevant in the age of Ingentic robotics. So I think that's an area that we're excited to make a bet on, right? So in one year, a lot of robots will be trained more in sim rather than collecting real-world data. So there's going to be a really strong reel-to-sim-to-reel flywheel. Hello, everyone, and welcome back to MTS. Today, we are joined by Neil Nhi, the CEO of Verne Robotics that just released V0.5, an open-weight world action model for industrial robots. So, Neil, welcome to MTS. Thank you so much, Sophia. I've been a huge fan of your show, and it's great to be here today.

[0:41](https://realalpha.app/v/zONxIHfiY9k/?t=41) What an honor. Well, it's great to have you as well. I will say, I love guests who bring props. And here I see maybe it's like a robot that's not connected to any intelligence right now. What's going on here? This is a data capture device that we use to help our customers capture data that they can use to train robots. So you can use it to, let's say, grasp some packages, do some assembly, right? And that data goes into our robot foundation model and can bring them autonomy. Robot Foundation model that is all open way and leading on some benchmarks. What can you share there? That's right. We recently released V0.5, which is our state-of-the-art robot foundation model.

[1:22](https://realalpha.app/v/zONxIHfiY9k/?t=82) It is currently one of the best U.S.-made open-source foundation models out there, and we're quite proud of that. We think that there is a huge future where we have to use open-source models to then accelerate the deployment of robotics. Yeah. I want to get your perspective because today is a very interesting day to talk about the future of open source, especially when it comes to the U.S. leading in open source. So what are your takes here? That's right. Well, we believe that a number of companies such as NVIDIA and Jensen specifically have really pioneered this idea of a U.S.-led open source effort.

[2:03](https://realalpha.app/v/zONxIHfiY9k/?t=123) everything from large language models that we are seeing to robotics, right? We think that it is not just in the interest of the U.S., but rather a national security question as well. And we think that having open source models not only lowers the barrier of entry for adoption for other companies, right, but also it provides, I think, a really strong technical moat where the U.S. can lead in this specific technology category. So, of course, NVIDIA was leading when it came to the U.S. being a leader in open-weight models, what do you think today's news might mean for the industry? Well, I think we would hope that open-weight models, whether it's from robotics to large language models, will still play a very important role in the ecosystem, right?

[2:48](https://realalpha.app/v/zONxIHfiY9k/?t=168) And we hope that stewards of these open-weight models will continue to push the frontier, right? But we do think that this puts us in a more precarious place. But I think it's hard to say what's going to happen next, But I'm confident that companies like NVIDIA and Hugging Face will continue to push the frontier. And we are very excited to continue to build on the platforms that they've created. Why precarious? I think there are a lot of competing interests here, as you know, right, Sophia? But once again, I think part of our job as a robotics company that is vertically integrated from models all the way down to deployment is that we're judged by the quality of the automation we can provide for our customers.

[3:29](https://realalpha.app/v/zONxIHfiY9k/?t=209) right so we want to you know be laser focused on that objective and that is fundamentally the north star that is driving the work that we do so a lot of people have said that gbt6 astra was like this gbt3 moment but for robotics because of what its capabilities were do you agree with that yeah i totally agree with that so i come from uh the research background so when the model was released a lot of my friends and I were texting in group chat. panicking right a lot of our graduate school research now went out of the window right uh so i mean it's it's both kind of unfortunate but also extremely positive for the community right uh so i guess the joke is that chat gpt is the gpt moment for robotics right uh so we think that uh agentic robotics will be the future and uh verne's model v 0.5 uh complements uh digantic future right we think that uh you know capabilities such as contact rich tasks capabilities such as full-body robot control for humanoids and semi-humanoids right are not fully there yet on the chat gpt level right so we believe that a system one model can really complement what a very smart and fast and reactive model like chat gpt can do right so we think that at least in the medium term both the system one and system two will coexist as very important

[4:51](https://realalpha.app/v/zONxIHfiY9k/?t=291) components. So why was GBT6 Astra such a breakthrough moment for robotics? What specifically

[4:59](https://realalpha.app/v/zONxIHfiY9k/?t=299) **Neil Nie:** around its architecture, its capabilities made it so? Yeah, great question, Sophia. I think the thing that surprised me and the community the most was the fact that it was able to learn how

[5:11](https://realalpha.app/v/zONxIHfiY9k/?t=311) to control a robot without actually being trained on a lot of robotics data. So this is an incredible new discovery and realization, right? This is almost saying that, you know, the bitter lesson holds true not just in large language models, but also in robotics. So this cross-domain knowledge transfer from 3D, whether it's 3D control or video games that transfer directly into robotics, right, was a very surprising, you know, in retrospect, not so surprising, but I think it was a watershed moment for us. How is that possible? Yeah, it's a very good question. I think, you know, without speculating too much on exactly the model is thinking, ChatGPT has learned a lot about 3D space and, you know, moving in 3D, right?

[5:58](https://realalpha.app/v/zONxIHfiY9k/?t=358) So that is very relevant to what it takes to control, you know, robots, whether it's humanoids or robotic arms. So now you guys have a leading open weights model when it comes to building agentic robotics. What are the advantages of having an open weight model? So we believe that open ways has a few key advantages, right? Number one is that it's very transparent. So we actually encourage folks to benchmark the model, try the model out, and we'll love to hear your feedback. And from our customer's point of view, right, they view that transparency as very important. I think the second thing is that we believe a lot of the competitive moat that comes from deploying robots actually comes from the customer relationships and the actual deployment itself.

[6:44](https://realalpha.app/v/zONxIHfiY9k/?t=404) And I think to accelerate that more, we need a better open source platform. So we think that another added benefit is that this can accelerate the development of the field, right? So we think robotics is not a zero-sum game. And I think the more players are deploying robots, the brighter the future is going to be. How come? Paint that picture for me. So, you know, we think that a lot of the best applications of these robots, especially in industrial use cases, are not just limited to these, like, really enormous factories, right? If we walk down the street in Fremont, California, just an hour away from here, you will see warehouses and factories that are really stretched thin because of how constrained their labor force is.

[7:26](https://realalpha.app/v/zONxIHfiY9k/?t=446) So we think that there's going to be a future where there will be a lot of Neo system integrators that are using these frontier open source models to then help these medium-sized businesses to accelerate and do more. And I think this is not just about saving costs, but also driving up their top line as well. Lovable also handles the backend and infrastructure, including managed Postgres, auth, storage, hosting, payments, and more, so you don't have to wire it all up yourself. And through Lovable's MCP server, you can create and deploy your project directly from the agents you already use. Turn ideas into software people love with Lovable.

[8:18](https://realalpha.app/v/zONxIHfiY9k/?t=498) Lovable.dev. Now, on to the episode. One of the things I wanted to ask about, what does this mean? I mean, right now, when we think about privacy, and being able to do everything locally. In the software world, of course, it's important, but not something crucial. In the robotics world, it's a make or break because you can't always go to the cloud. If there's a little bit of latency and you're running a full humanoid and there's not great Wi-Fi in a corner of the house and a humanoid breaks, that kills the entire product. And so does that, paired with open source, have a big factor as well? Yeah, totally.

[8:59](https://realalpha.app/v/zONxIHfiY9k/?t=539) We think that these System 0 and System 1 models are meant to be running locally on device, right? And we think that that's exactly how we can solve some of the problems you mentioned, right? Things like latency, right? What if the Wi-Fi drops and things like that? But these, like, bigger models like Astra can still run on the cloud because they don't have to run a couple times per second, right? They can run every second or even slower than that. So we think that an open source model can allow people to optimize that and then install it and run it locally on-device, right? So we think that on-device inference is also going to be very important going forward.

[9:34](https://realalpha.app/v/zONxIHfiY9k/?t=574) So explain it to me, for people who aren't familiar with the robotic space, the hardware space, how does on-device inference work? Why is that super important for robotics? Yeah, 100%. So these modern-day humanoids that you're seeing have, you know, almost half a dozen cameras, whether it's mounted on their head, on their, you know, end effectors, on their grippers, right? And these images get streamed to this really massive model, oftentimes almost 10 billion, you know, parameters, right? That's living on the GPU on the robot. And this model has to run almost as fast as the human reaction time, right?

[10:10](https://realalpha.app/v/zONxIHfiY9k/?t=610) **Neil Nie:** So fun fact, if you play baseball, right, you know the human reaction time is about 200 milliseconds, right?

[10:15](https://realalpha.app/v/zONxIHfiY9k/?t=615) So that means that this model has to take all of these images and produce actions for the robot five times per second, right? So that is very fast, and internet latency sometimes can produce a lot of jitter, right? So when it's a safety-critical system, right, we believe that running it locally can have a lot of benefits, right? So this is kind of a delicate balance between doing that locally plus the added intelligence of a GPT-Astra, right? Is that the industry standard right now where most robots are being run locally? We do see a lot of that today, right? So I was actually just at the conference on robot learning in Austin, Texas, right?

[10:53](https://realalpha.app/v/zONxIHfiY9k/?t=653) When we saw a lot of the demos there, I would say probably half of the company were running, you know, policies locally, right? Whether it's at a desktop, in a booth, right, or just directly on the robot. And when they're running these locally, what models are they typically running? Yeah, good question. So a lot of companies like DeepMind, obviously, right, they have their own leading models. A lot of startups are fine-tuning Pi 0.5, right? So Pi's frontier open source model. But now we are reaching out to more and more partners, especially startups, right, to see if they want to try V0.5 as an alternative, right?

[11:28](https://realalpha.app/v/zONxIHfiY9k/?t=688) It's a fantastic world action model. And we have seen great results fine-tuning that and serving that for different tasks. What are some of those results? Yeah, so we are particularly proud of full-body tasks that we've done, right? So imagine like if you have a robot that is supposed to stock shelves in a grocery store, right? Like you have to bend down and pick up some inventory, put it on the shelf and then turn around, right? So those type of full body tasks we are quite excited about. And I think that also illustrates an area where whether it's. PAI GPT, Astro, or PAI 0.5 are not as particularly good at, right?

[12:04](https://realalpha.app/v/zONxIHfiY9k/?t=724) Who are the current leaders when it comes to local models that are being used in robotics? So we think that NVIDIA Groot is a really fantastic model. They can run fast locally on device. Props to physical intelligence, right? PAI's model, especially their last year's open source model, is also quite strong, right? And I think those are, you know, the two top of mind for us, right? And a lot of the other ones we are just not as aware of, you know, because they're not open source or they're very private or stealthy. But where are they? So compare that to the capabilities of GPT-6. What's the gap there?

[12:44](https://realalpha.app/v/zONxIHfiY9k/?t=764) So on some of these benchmarks, right? So that's a really good question. I think GPT-6 definitely excels at more general tasks, right? So if you are given the robot some previously unseen environment, ChatGPT can do better. But if you are, let's say, strictly benchmarking for cycle time, speed, or other metrics like that, then these special purpose-made robotics models can still win. But we don't think that that's necessarily going to be the long-term future. We think that ChatGPT will also be able to absorb some of those tasks as well. So do you think the future is a mix, a hybrid? So humanoids, I'm just using humanoids as a robot example here, will have a local model that can respond instantly no matter what, even with no Wi-Fi.

[13:30](https://realalpha.app/v/zONxIHfiY9k/?t=810) But then for more difficult tasks and, say, it assesses its situation and says, okay, it's fine if I have a little bit of a lag time, then it goes to a more advanced model in the cloud. Yeah, 100%. I think your intuition there is really spot on. And I think more robotic companies are starting to adopt that formulation here. Yeah. I know you say the U.S., especially NVIDIA, was leading the frontier and the charge for having more open-weight models. We're hearing this year is the year that the capabilities are good enough for robotics that we're seeing our chat GPT moment, but in the physical world.

[14:06](https://realalpha.app/v/zONxIHfiY9k/?t=846) given that maybe NVIDIA who knows i'm speculating here i have no idea i'm not sharing any opinions but say NVIDIA isn't able to lead in this domain anymore who do you think will be the leader when it comes to local open weight models that can be purpose-built used used in robotics so i think that's that's a really good question and we want to make sure that Vern is well-positioned to offer a really great open weights platform, right? So we want to be one of the companies that you can lean on to use these kind of open weights local models that you can run inference on the edge, right? And so we definitely hope that, you know, we will be able to commit resources and train and release these models so that we can serve the community and accelerate the deployment of robots. But I think also we are excited to see if other companies that have released models in the past, like Pi, will also join forth and see if they will release models too.

[15:08](https://realalpha.app/v/zONxIHfiY9k/?t=908) But I think right now, NVIDIA is still definitely the category leader here. Yeah. I mean, we had Runway on yesterday that was talking about Praxis One. I think there's a very interesting idea around people who've been doing video who now have a lot of that rich data set to be able to build strong world action models. And so I wanted to get your take on, And do you think people who've been generating video will be a frontier leader when it comes to creating great world action models? Well, we believe that to solve a lot of these challenging problems in robotics, we need to lean on many different aspects of machine learning, right?

[15:49](https://realalpha.app/v/zONxIHfiY9k/?t=949) But it's still very much a robotics-driven problem. So I think we particularly pride ourselves on that, right? Our team comes from that really amazing background. but I think what is Really encouraging, Sophia, by the way, is seeing so many players now switching over, right? So it's almost as if like Groot and Vern, like we were pretty early in the discussion, right? And then now we're seeing companies like Runway that you mentioned also doing this Black Forest Lab. So another company that came out of Stanford SVL, which is my old lab, right? So, you know, it is a small world and I think it's very encouraging to see these open-waist efforts, right?

[16:25](https://realalpha.app/v/zONxIHfiY9k/?t=985) But we want to make sure that our model remains commercially available so you can deploy them and train them. Right. And I think, you know, just like LLM's competition will only accelerate how good these open waste models will become. And for Verve 0.5, what's the unique edge? Yeah, good question. So we think there are a few clear advantages. Right. So the first is that it's a full body control model, which means that if you have, let's say, a semi-humanoid solution where you need to move around in a factory, you know, where we're in a warehouse. Right. And, you know, to be honest, most tasks are like that.

[16:58](https://realalpha.app/v/zONxIHfiY9k/?t=1018) You can lean on us and we will be able to produce actions and help you automate those tasks. So that's number one. Number two is that the model was trained to work in an agentic system, actually.

[17:10](https://realalpha.app/v/zONxIHfiY9k/?t=1030) **Neil Nie:** So the model actually has an internal understanding of if a subtask or a task has been completed,

[17:16](https://realalpha.app/v/zONxIHfiY9k/?t=1036) so then ChatGPT Astra can step in and switch to the next step of a task. So it's trained and meant to be used with these more agentic systems. So we talked a lot about capabilities that these open weight models will be able to have when it comes to robotics. Wanted to get your prediction. Anything you think will be true in a year from now that isn't true today? Well, that's a great question. So this is partially inspired by the stuff that we discussed with some of my old colleagues at the conference. One is going to be simulation will become more and more relevant in the age of energetic robotics.

[17:50](https://realalpha.app/v/zONxIHfiY9k/?t=1070) So I think that's an area that we're excited to make a bet on, right? So in one year, a lot of robots will be trained more in sim rather than collecting real-world data. So there's going to be a really strong real-to-sim-to-real flywheel, right? So maybe we can check in in a year and see if that's going to be true. Man, we'll have to test it out. Last question I have for you for people who are just following along. What is the latest with Avern? What can people expect to see over the next few months? Great question. So we just released our new open-ways model, And we're excited to launch a few more products in the coming month.

[18:26](https://realalpha.app/v/zONxIHfiY9k/?t=1106) I won't mention exactly what they are, but stay tuned. We're very excited. And we are also growing and expanding the team all the way across engineering to deployment. So if you're interested, please visit our website. And I think 2026 will be the year of models and a lot of this research, right? And 2027 will be the year of deployment. So we're going to see more and more companies deploying robots. We at Vern are already actively deploying with a number of our customers, and we can't wait to triple that in the next coming quarters. Amazing. Well, Neil, I really wanted to thank you for joining us today on MTS and sharing about the latest updates with Vern and what this means for the future of the robotics industry.

[19:14](https://realalpha.app/v/zONxIHfiY9k/?t=1154) Thank you for joining us. Thank you, Sophia. It's such a pleasure to be here. Such a pleasure. A huge thanks to MTS sponsors. security and governance. KongHQ.com. Finally, prove what your AI investment is worth with Lairdyn. Head to Lairdyn.com today to start measuring your AI's bottom line. DB, trusted to run critical applications by 75% of the Fortune 100, the intelligent data platform for the era. Apply today for access to up to $15,000 in Atlas credits, dedicated technical support, and exclusive events and go-to-market opportunities. MongoDB.com. Chargebee, billing and monetization infrastructure for the AI era. chargebee.com.

[20:24](https://realalpha.app/v/zONxIHfiY9k/?t=1224) ClickHouse is the real-time database built to keep up with every query your apps and agents send its way and it now runs postgres observability and ai tooling on one platform learn more at clickhouse.com you ship fast so do attackers don't assume you're secure prove it horizon 3 go hack yourself with ai visit horizon3.ai slash node zero blitzy autonomous software development for enterprise code bases ship 5x faster blitzy.com adquick make your brand a billboard out of home advertising as easy to scale as digital adquick.com arena measuring ai performance in the real world arena.ai support for the show comes from vcx the public ticker for private tech the u.s stock market started history's greatest wave of wealth creation from factory workers in detroit to farmers in omaha anyone could own a piece of the great american companies but today our most innovative companies are staying private longer, which means everyday Americans are missing out. Until now. Introducing VCX, a public ticker for private tech. Visit getvcx.com for more info. That's getvcx.com.

[21:39](https://realalpha.app/v/zONxIHfiY9k/?t=1299) Carefully consider the investment material before investing, including objectives, risk, charges, and expenses. This and other information can be found in the innovations fund prospectus at getvcx.com. This is a paid sponsorship. You
