Try Pro free

The best 12 min of 132

Robotic Deep-Dive: Beating China, the Race Against Tesla, & New Robotics Timeline With Brett Adcock

Brett Adcock says the humanoid race hinges on integrated hardware, AI and manufacturing—and that useful robots for homes are getting closer.

Moonshots · Peter Diamandis & Dave Blundin

The best minutes

  1. 16:19China's robot dance demos do not prove practical autonomyPro pick 3 min
  2. 24:06Figure runs its robot intelligence on board to meet latency needs5 min
  3. 1:01:00Figure expects lights-out robot manufacturing within five years4 min

Every moment plays free. PRO marks a write-up that's in Pro.

Also worth watching

  1. 3:19Figure will wait for a high-quality home robot before launch6 min
  2. 12:18Figure’s coffee-cup demo helped validate neural nets in robotics4 min
  3. 36:48Figure built its own actuator after a supplier’s part was too large3 min
  4. 1:12:11Figure estimates Baku capacity at over 40,000 robots a year3 min
  5. 1:43:27Compute bids turned against Figure as GPU supply tightened5 min

Skip the founder profile and Silicon Valley discussion, the AI-device discussion, the health segment, and the sponsor reads.

Full analysis · 7 min read

Real Alpha · Nov 16

Brett Adcock says Figure 4 could be humanoids’ iPhone moment

  • Adcock says Baku could make maybe over 40,000 or so robots yearly at full rate across four lines.
  • That is stated capacity, not shipments, and his autonomy evidence lacks published success or intervention rates.
  • For builders, the test is whether home performance, compute, and manufacturing scale together.

Free with an account

Read the rest of the analysis

7 sections, the 5 other voices it tests Adcock against, and what it means for you. Every episode we cut gets one.

Signed up before? The same buttons sign you in. The episode, its moments and the transcript stay free.

16:47

Adcock makes practical autonomy the real race

Adcock’s four-stage frame runs from functional hardware and an AI recipe to scaling intelligence and robot production. He measures progress by useful work in unfamiliar settings, not choreography.

At 16:47, he describes Chinese dance demonstrations as open-loop systems, often trained from human motion capture. He says they do not react to the environment.

That distinction matters because movement can look autonomous while replaying a prepared sequence. Still, the episode gives no matched tests across companies, tasks, conditions, or success measures.

Adcock runs Figure and names China as its closest competitor, making this a founder’s assessment rather than an independent comparison. A host later disclosed investing in Figure, relevant context that neither proves nor disproves the technical account.

Josh Gruenstein, Tutor Intelligence’s CEO, said on MTS in November 2026 that robotics companies must develop models, hardware, and deployments together. That supports the whole-stack framing, not Figure’s lead.

“they don't react to the environment at all.”
17:16

Our read

The episode draws a useful line between movement and autonomy, but does not establish that Figure is ahead. The test is useful work under changing conditions, measured against comparable alternatives.

25:40

Helix keeps fast control onboard the robot

Adcock says Helix runs onboard, from planning and reasoning through low-level trajectories and torques. Cloud services handle speech and fleet-level coordination, such as swapping robots on a production line.

This split follows two clocks. Body control operates at millisecond speed, while fleet instructions can tolerate slower updates.

At 27:31, Adcock says network delays make cloud control unsuitable for fast dynamics. He gives 20 to 30 milliseconds each way over strong 5G, nearly 50 milliseconds round trip before inference.

That is an engineering constraint, not a measured performance result. The episode supplies no latency tests or outage benchmarks.

Neil Nie, Verne Robotics CEO, made a similar case on MTS in November 2026. He separates fast, safety-critical local policies from slower cloud reasoning.

For example, losing internet should not stop a robot mid-laundry, as Adcock says onboard execution permits. But reliability also depends on stops, recovery, and observed failure rates.

“We do all inference on the robots you're seeing here on board today.”
25:40

Our read

The local-control case follows from latency and connectivity constraints. The episode explains the architecture, not how reliably it handles faults in practice.

23:21

Unfamiliar-home tests still fall short of launch proof

Figure’s home case has a more concrete basis than a dance demo. Adcock says a robot worked in 30 rental homes it had never seen.

The tasks included making beds and tidying living rooms without prior exposure to those homes or objects. That supports a form of generalization, not reliable household autonomy.

For example, one successful tidy-up does not establish a dozen-hour shift, daily use, or recovery when children interrupt.

Adcock says current testing covers roughly 20 to 100 new California homes at a time. He hopes to show large, long-horizon tasks in any home within six months, perhaps sooner, but says he cannot promise it.

Josh Gao, Mecka AI’s CEO, discussed deployment failures and corrections on Training Data in October 2026. He sees those as inputs to better robot systems.

That comparison makes missing operational data important. Figure provides no task-completion, intervention, uptime, or damage rates here.

Until such figures appear, these trials show capability in selected tasks, not a launch-ready service. The next evidence should show performance across homes and longer work sessions.

Our read

The home trials make generalization more concrete than a staged demonstration. Their commercial meaning depends on performance data across varied homes and longer tasks.

34:49

Figure 4 raises a hardware claim and burden

Figure 4 carries Adcock’s largest hardware promise, but it remains a forecast. He says the robot should cost less and do more.

After seven hand iterations, he offers an ambitious comparison between the new hand and a person wearing a glove. That is not a dexterity benchmark or proof of household performance.

An earlier integration story shows why component design matters. Figure’s first sourced actuator met specifications but was three times too large and three times overweight.

Adcock says the company built its own after that part proved too big, heavy, and costly. Bringing an actuator in-house also required work on firmware, thermals, motors, gears, wiring, and packaging.

Sergey Levine, Physical Intelligence cofounder, said on Minus One in October 2026 that robotics failures must be traced across hardware, software, and models. His point fits the integration burden, not the Figure 4 claim itself.

Still, vertical integration solves a fit problem by creating more engineering work. Figure has not shown Figure 4’s price, lifetime, or production yield.

“i would make the argument that like a figure four with a glove and a human with a glove is almost indistinguishable”
34:49

Our read

The actuator story explains why Figure builds critical parts in-house. The Figure 4 hand remains a forecast until capability, durability, and production costs are demonstrated.

1:12:37

Baku’s stated capacity is not production

Adcock says Baku could make maybe over 40,000 or so robots a year at full rate across four lines. That is stated capacity, not current shipments.

The episode gives no actual output, utilization, manufacturing yield, or cost per unit. Adcock says a million-unit facility may be possible if Figure solves its next scaling problems.

That condition matters. Intelligence, data, compute, supply chains, and integration all constrain the path to higher output.

At 1:04:30, Adcock says full lights-out manufacturing feels achievable within five years, while the nearer timing remains foggy. He says Figure 4 was designed with that goal in mind.

Take the distinction seriously. A robot operating a machine is not the same as a validated factory producing robots without human labor.

He also describes a zero-shot deployment at a client’s warehouse, but not a fully automated production process. Factory plans remain an ambition backed by rated capacity, not proof of self-replication or unit economics.

Our read

Baku’s capacity estimate makes the manufacturing ambition tangible, but it is not achieved throughput. The episode gives a target and a conditional forecast, not a demonstrated cost curve.

1:44:13

Compute shortages could slow the training plan

Adcock says Figure spends low nine figures annually on compute, much of it for training. He expects billion-dollar Helix training runs next summer, if the pieces come together.

That plan requires usable clusters, prepared physical-world data, and the ability to bring both online together. The episode describes an N-Scale deal for 100,000 GPUs and more than $3 billion, with capacity expected in the second half of 2027.

Adcock confirms that timing, but contracted capacity is not delivered training capacity. He recounts getting no bids when seeking more, then receiving one costly offer.

That is evidence of Figure’s buying experience, not a complete survey of compute supply. The distinction matters when a training plan depends on capacity that is not yet online.

Rohan Anil, Core Automation cofounder, discussed training across fragmented GPU sites on MTS in November 2026. He said suitable algorithms could unlock scattered capacity.

That is a possible route, not evidence Figure can use it. Meanwhile, Adcock says physical data must be audited, cleaned, and annotated before training.

So more GPUs do not instantly translate into better models. Capacity, setup, and data readiness all determine what Figure can actually train.

“I think, hopefully, I expect that next summer we'll be doing billion-dollar training runs on Helix.”
1:44:13

Our read

The compute plan is specific, but depends on delivery, data readiness, and cluster setup. A GPU count alone does not show how much training capacity Figure can use.

4:50

Home sales wait on reliability and service

Adcock says Figure will wait to sell a home robot until it can work without teleoperation, failures, or stopping because it is unsure what to do. His launch bar is higher than a demonstration.

Commercial tasks may be specified, but homes vary in layout and requests. For example, folding laundry, unloading groceries, and handling dishes require different interactions with objects and surroundings.

Adcock says Figure plans to handle early home maintenance itself to control the experience. That choice brings a service burden alongside the hardware and software work.

He says affordability will require some sort of $300 to $500 monthly lease. That is a rough requirement, not an announced price or financing plan.

Safety requires physical safeguards, semantic judgment, cybersecurity, and testing, he says. Still, the episode provides no performance rates, repair terms, or consumer safety record.

Leasing could make upgrades and recurring support possible, but the economics depend on dependable work. A monthly figure matters less than what the robot completes and how often people must intervene.

Our read

Waiting protects the product’s reputation, but leaves timing and household economics open. The milestones to watch are reliable work, safety evidence, and a service model that holds up outside the lab.

What it means for you

  1. Track task reliability. For any robot pilot, measure completion rates, human interventions, task duration, and failures across unfamiliar environments. Figure’s home trials make those measures more informative than a highlight reel.
  2. Test local control. Map which decisions need onboard responses and which can tolerate cloud delays. Measure behavior during network loss, not only on a stable connection.
  3. Separate capacity from output. Ask for actual units produced, utilization, yield, and cost alongside rated capacity. Baku’s stated full-rate capacity is not evidence of current shipments.
  4. Verify compute availability. Separate GPU agreements from delivered, usable capacity, and track when training data is ready. Figure’s account shows both supply and data preparation can constrain model runs.

What to watch

  • Whether Figure publishes task-completion and intervention rates for unfamiliar homes.
  • Whether the N-Scale capacity comes online in the second half of 2027.
  • Whether lights-out manufacturing arrives within Adcock’s five-year estimate.

Send this to product and operations leaders assessing humanoid-robot pilots.

32:40

Watch the minute where Adcock frames Figure 4 as a possible iPhone moment.

Written by Real Alpha from the episode's transcript and the other moments it links, then checked against the transcript. Quotes are the speaker's exact words.

Minute by minute

  1. 3:19Home robots must work reliably

    Adcock distinguishes commercial work, with specified tasks and demanding performance goals, from homes, where layouts and tasks vary. He says Figure will wait to sell a home robot until it can work autonomously and reliably for long stretches. He hopes to demonstrate long-horizon work in any home within six months.

  2. 11:02The humanoid race has four stages

    Adcock lays out functional, affordable hardware; an end-to-end AI recipe; then scaling intelligence and robot production. He traces Figure’s progress from a neural-network coffee-making demonstration to full-body control, saying walking and hand movements must be trained together. He argues that many Chinese robot dance demonstrations use open-loop replay rather than reacting to the world.

  3. 19:48Helix targets real-world control

    Adcock says many general-model robot demonstrations use fixed cameras, basic tasks and slow movements, unlike the fast, full-body work Figure targets. He says Helix 2.5 handled tasks in 30 rental homes without prior exposure to the homes or objects. Figure runs Helix on the robots and has a deal for 100,000 GPUs costing three-plus billion dollars, scheduled for the second half of 2027.

  4. 29:15Data scaling points to Figure 4

    Adcock says Helix 2.5 showed model loss falling and task execution improving as data increased, and he characterized early Helix 3 as a childlike learner. He says home robots may need a $300-$500 monthly lease; Figure 4 aims to cut costs and improve capability, especially through its hands. He says Figure builds components itself after outside actuators proved too large, heavy and expensive.

  5. 41:53Robot design must earn trust

    Adcock says Figure is working on head and face design, gestures and persistent memory to make robots easier to interact with and trust. He says a user’s preferences and memory should transfer between robots, and people will be able to choose apparel and voice. He sees a home robot as both a useful worker and a possible companion.

  6. 49:16Adcock treats safety as a full-stack problem

    Adcock says safety spans physical failures and judgment, backed by system design, testing and cybersecurity; he says there is no single fix. He says Figure is moving component sourcing away from China amid tariffs and geopolitical concerns, and sees space as a potential use but says the company is not ready and will avoid defense. He says Figure will not train robots to defend themselves from people.

  7. 1:01:00Adcock sees a path to lights-out factories

    Adcock says Figure 4 was designed for robots to control machines that build robots, enabling lights-out manufacturing. He sees full lights-out production within five years but says he cannot say whether it will happen in months or years. He describes the pace of progress as much faster than he expected when he started Figure.

  8. 1:08:48Robot skills should generalize

    Adcock says diverse physical-world data should help robots adapt to new tasks with limited instruction instead of needing narrow task-specific training. He estimates Baku can produce over 40,000 units a year at full rate and sees a difficult path to one million a year at a facility if Figure solves current scaling problems. He also imagines demand reaching roughly 20 billion humanoids.

  9. 1:20:19Chip access may limit scale

    Adcock says current robot chips remain too costly and power-hungry, while memory partnerships support near-term expansion. He is wary of designing custom silicon years ahead because Helix’s compute needs and model architecture have changed quickly. He says Figure will need access to unprecedented amounts of silicon over time.

  10. 1:27:14Adcock calls China Figure’s closest competitor

    Adcock calls collective China Figure’s closest competitor, while saying Chinese companies are pursuing different goals and need better hardware and policies. He says the Bay Area has the talent density to build general-purpose robots and recounts moving there after Vettery and Archer. He says Archer went public to raise needed capital, though private ownership would have been healthier, and expects Figure to go public eventually.

  11. 1:41:29Compute supply is tightening

    Adcock says Figure has a large backlog of physical-world data that must be audited and prepared before training. He reports low nine-figure annual compute spending and expects billion-dollar Helix training runs next summer if the capacity comes together. He says bids for additional compute dried up by September 2026.

  12. 1:48:07Hark targets personalized intelligence

    Adcock says Hark is developing personalized AI across computer use, speech, memory and vision, with Hark Pro as its software product and devices in development. He says Hark had over 1.2 million signups and almost 5 million browser sessions after about a month, and that its speech model also powers Figure robots. He says privacy is central because the system would hold personal information.

  13. 1:57:00AI may replace today’s devices

    Adcock expects AI devices to replace phones and computers as the main interface, with watches, glasses and earbuds serving as accessories. He says Hark has hardware in its lab and sees brain-computer interfaces as a later route for AI interaction. He argues that AI models and devices should be designed together.

  14. 2:05:03Adcock sees little robotics regulation

    Adcock says Figure has seen little public discussion of humanoid regulation beyond the FCC ban on Chinese humanoids. He says Figure plans to ship only when its products are safe and reliable. He describes building companies as a major opportunity for the next generation, while saying that running three companies alongside raising three children has been exhausting and rewarding.

Published Nov 16, 2026.

Transcript · 1 h 53 m read

Moonshots, hosted by Peter Diamandis and Dave Blundin, with Brett Adcock.

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

  1. As you think about competition, is it Tesla and China? Is that the way you see it in the world right now? I personally haven't seen it much out of Tesla in two or three years. I've seen China do a lot of dancing. We care about something very different than what we think China cares about. We just got to go beat them. It's a real race here. The work isn't really practical. The technology stacked and needed to get there isn't practical. And the AI stack to do it can be done by a senior in college. My view is there's like roughly four stages for this humanoid race. My chapter one of this book is you have to build like really good functional hardware.

  2. Two is you have to be able to build an AI recipe. Chapter three and four for me are like scaling intelligence and scaling robots. What might we see with figure four? What areas are you focusing on making improvements? We're like in the flip phone era for humanoids. Figure four is probably like iPhone one moment we're going to have here soon. We need the robot to get cheaper. We're trying to make the robot more capable. A little sneak peek of that is we've been working on the hands. A figure four with a glove and a human with a glove is almost indistinguishable. That's like an iPhone moment. Figure 3 is probably the best humanoid in the market by far, but it'll make Figure 3 look like a caveman.

  3. Now that's a moonshot, ladies and gentlemen. This episode is brought to you by The Abundant Summit and Link Ventures. So, Brett, good to be here for our annual catch-up. It's been about a year since Dave and I were here to see Figure, and quite the difference. to walk in, you just park in the parking lot over there, walk in, and there's like a hundred robots outside to greet you. So, my question is, what do you call a group of robots? I'm calling them a flock of figures. Do you have a name for them? Those are our visitor bots. They walk around all of campus. They can open doors, walk on any stairs, use the elevators.

Read the whole transcript
  1. And they're there to greet people, talk. And we're right now working on trying to get them to get coffee for folks. We have a kind of a stretch goal of like, it'd be great if you can come to campus and never talk to a human as you get set up. You know, get fully checked in, walk to the conference room, do a tour, like go to the bathroom, like all done through a humanoid. Let me show you where I was built. Let me show you what I'm doing all day. Yeah, I know, exactly. Like we do that all the time. So like, why not? We should have a humanoid. And you said something a minute ago that's pretty profound.

  2. You have more figure robots than humans now. Yeah, we have more figure robots on campus. When did that happen? probably like four or five months ago we uh crossed the number of i think we have like seven eight hundred people here so we crossed the number of uh yeah we have more robots now we have many more now than humans and you're up to how many produced at this point I don't know, like 1,500 or so on campus, something like that, in order of like, we have well over 1,000 on campus, and we have robots out in other places in the world as well. Oh, nice. So we're seeing figure three around here.

  3. Last time we were here, you were just transitioning to figure three, I think, at that point. How long before voice, how long before I'm talking to the figure? You can talk to any robot here. Yeah? Yeah. You can stop it and just say, ask your questions. You can talk to it. All right, I'll have to do that. It's got real-time speech-to-speech, yeah. Yeah, so every robot here is equipped with speaker mic. We actually use Hark for our voice AI. And every robot here has got, like, full real-time speech-to-speech. Incredible. Yeah. Incredible. So, you know, a lot of people are asking me, so when am I going to be able to get my –

  4. Brett Adcock · 6 min

    “But our view here is that Figure will ship a robot when it ships a really high-quality product in the home.”

  5. You know, do you think you're going to keep robot maintenance or repair as a figure in-house capability? Or are you going to sort of approve outside vendors to do that all over the world?

  6. For now, we're going to do it ourselves. We really want to make sure the robot has the highest quality experience in the home. And we know the robot's better than anybody else. We built him. So we're going to do that ourselves. It can just crawl into a box and ship itself back. Yep, totally. Put the label on itself and the hand comes out, slaps it on, and closes the box. There's many ways to solve this problem. The end state is for sure us not doing everything. But in the beginning, as we ramped the product up, Like, you really want this incredible experience, like, in the home. Like, it's basically just like another human is in your house helping you out.

  7. We're only going to get that with, like, the highest level of care and attention. And we're putting everything, our blood, sweat, and tears, like, all into this to make sure, like, it comes out really rocking. The flip side, if it comes out and doesn't work really well, you're kind of dead. So, like, so we really want to make sure it's a really, really high confidence. Were you a Steve Jobs fan as a kid? Of course. Yeah, right. Steve is great. The parallels are unbelievable. You know, I was with Mike Rowe and Joe Polish at an event recently. And Mike Rowe was the guy who did the Dirty Jobs show.

  8. Yeah. And he was, I told him I was going to be coming here. And he asked, when are we going to see, you know, figure robots doing, you know, welding and roof repairs and sort of all the blue-collar work? You know, there's this been interesting conversation that AI is coming for white-collar jobs right now. And it is from finance and legal and diagnostics. Yeah. But the question of when will these robots start coming into vendors doing traditional bricklaying and all the manual labor out there? I mean, my thesis here – let me lay out my thesis for robots. I feel like it may be helpful to kind of give you guys my sense.

  9. My view is there's like a high level, like roughly four stages for this humanoid race. My chapter one of this book is you have to build like really good functional hardware. The robot hardware is going to be incredible. You can't build AGI on a robot without really good hardware. And we're doing that, obviously, with now figure one, two, and three. We have four coming out at some point. We think we're like, you know. But anyway, the hardware needs to be able to do the speed, torques, necessary that a human could with the same range of motions. Roughly at a high level. It could be affordable, and you've got to make a lot of them.

  10. So that's one. Two is you have to be able to build an AI recipe. that can run end-to-end longer highs and work. Full loco manipulation, meaning like walking the body and moving the hands. And you really haven't, besides figure, really haven't seen any of that in the world. You've seen like groups do like tabletop work with like off-the-shelf robots and other stuff that are like bolted to a table that are overfit to a certain environment. You haven't seen robots be able to like go off autonomously, be able to walk around and do stuff. And we started doing that with Helix too. Like full, like full, like full, like room-wide scale.

  11. Brett Adcock · 4 min

    “And we did it, and I think we raised like a little under $700 million, like shortly thereafter.”

  12. Brett Adcock · 3 min

    “All of this work is completely orthogonal and different than what you need to solve general robotics.”

  13. stop since we did that four months ago. They run practically every day. Somewhere in the AI. They go curing themselves, too. But could you please let me do something else? But you're hearing an important part. Chapter three and four for me are scaling intelligence and scaling robots. And what we really want to see is what you're mentioning here with the submarine. We want to see as you're putting more input in, you're getting better output.

  14. And for us, our input is data and compute, and our output is getting robots to do more useful work. Well, you're blowing through it. So stage three is data, data, data. It's data and compute. It's data and compute. It's AI scaling laws. when you stood up helix you needed to the models that existed back then were not interfacing with physical systems and Then, recently, you saw GPT-Astra actually be able to operate robots. So talk to us about that. How far ahead is Helix from a general model that can operate a physical system? I would say one thing I'll say for these other type of language models, and even multimodal models we're seeing, command robots, they're all being done for very basic tasks on a fixed camera frame very slowly. If you slow robotics down enough, none of it becomes a really difficult autonomy robotics problem, in the limit. These videos are shot with 5x, 10x, 20x speeds on a very slow...

  15. That's true. People realize they're sped up. They're sped up because they slowed everything down. You're basically building closed-loop control, but doing it really slow. They need that feedback, but the models are not good enough to output this really quickly. They're very low-dimensional spaces on very basic tasks. We just have a different problem to solve. We need to move at very different dynamics and frequencies than what we're seeing out in the world in these videos. We need a robot to walk around. It needs to be able to react to the environment really quickly. It needs to be able to move a package over there at sub-three seconds of package. It can't really mess up very much, only a few percentage points, basically, over a whole shift. We need robots to go into home, do a very delicate fine-grained manipulation, grabbing cups and dishware and put them away and open doors and cabinets and navigating through tight spaces like a bedroom or kitchen. Then we have to do that on maybe one of the most complex mechanical machinery ever built, and not mess up. You have to balance on a quarter-sized point in your foot above the center of mass in your torso. It becomes a very different problem

  16. when you attack it like this than what we'll see on tabletop college setups. You're seeing that now with Astra and the rest of the groups. They're not real models for robotics. If they were, we could run them. There's another subtlety in there. Your entire AI stack is totally owned by either Figure or Hark. When you do a university setup, tabletop, whatever, it's on top of at least one layer of third-party neural net LLM. It's just not real. You can't put them into really high-dynamic robots like this and make them work. You can't scale them. There's no path that we can do, like, the Keurig K-Cup, go from there to full local manipulation to go to, like, there, like, to what we did with...

  17. like Helix 2.5 recently. The architecture needs to dramatically change. These are setups that have basically unlimited compute. They're plugged in with really low latency comms through a network. They have fixed cameras. They're overfitting the tables, generally, for data collection. There are not any scenarios that are realistic, what we'll see. So, we had to go build Helix from the ground up to build, how do we build the right brain for the robot. Right now, I would say, arguably, we understand how the brain works really well now. What the brain is starved of today is data and compute. We don't have enough world understanding of what forces feel like in joints and observations to be able to get it into every home in the world. But we do see a path now. With Helix 2.5, we went into 30 rental homes that we rented that day or day before. And we were able to do work, like make the bed and tidy the living room without ever seeing the house or objects before.

  18. It was like fully zero shot. They call it zero shot in the air community. None of this was in training. That was the first time things were about seeing this. Think of a bunch of stair steps that we've been taking now over the last three years that are all building towards, how do we get that now to go from there to drop into any home for 10 hours and it can do work? And that's the trajectory we're on now. So, that really gets me to the heart of your question, is when we're going to see that. We've arguably seen some of that this year in 2026. I think you'll see more from us here, even this year. And then, I'd

  19. Brett Adcock · 5 min

    “So, everything from the planning and the reasoning to even the low-level sending trajectories out and torques, all that's done on the robot today.”

  20. Let's talk about training. World models versus physical versus, you know, what you've been doing in terms of task, video recording. What's your vision on training to get to the level of fidelity you need? We took a pretty big bet on Helix 3 here, which we talked a little bit about on the all-hands, but we haven't released anything publicly yet. If you roll back before that, we released Helix 2.5 about a few months ago, where we basically showed zero-shot reasoning in 30 unseen homes before. before. We also showed what we think is the first humanoid scaling laws for data as we increased data, we saw a very log-linear relationship, meaning model losses were coming down and the robots were actually executing better in these environments with more data than with less data, which is unbelievable to see. We've seen this in LLMs. We haven't seen this in robotics before then. If this trend holds, then as we're adding more compute and data to the problem, the robots are literally going to get smarter. They're going to get better at the TAS every single day.

  21. We have a new work we've been doing on Helix 3. I think we gave you guys a little preview today. And I would largely characterize this as little baby AGI. We have a dude in a suit in the office. We have a secret lab called Black Box you guys were in, which is a little home. And you said we're the only ones that have ever, outsiders that have been in there? Yeah. Wow. That's great. It's a mind-blowing experience because when the LLMs were improving, they didn't really look anything like a child. Like they were doing complex math, but they were also hallucinating. It's nothing like a kid. But what we just saw in the black box looks exactly like a kid learning how to do things.

  22. And it's completely, you know, nothing is pre-programmed. It's all neural net. You ask it to do a task it's never heard before. And it looks exactly like a kid who's learning how to do it and experimenting and trying and rotating and, you know, pick up the chair and like, oh, it's not going to work. And you can just visualize how this is evolving just like a kid. It's so much cooler than the way the LLMs evolved. So your production rate right now, I mean, the demand for these robots in homes, you know, I'm asking people all the time if these things are costing you and are you still tracking towards a $30,000 price?

  23. tag, $300 a month type of lease rate? Yeah. In the home, we'll have to make the robot really affordable. You'll need some sort of $300 to $500 a month lease. $10 a day. It's under $1 an hour. I ask people when I'm on stage, how many of these would you own if it was that? And the average individual is at least two. Yeah. You've got to either lease it or buy it. There's only two ways we've got to get the robot out. Do you have a preference? Not really. I think through Elyse, I think you can really bring the cost down quite a lot and help amortize that better for the customer. I don't really know if in...

  24. And I think the technology would be radically improving. My view is that we're like in the flip phone era for humanoids. I think I would make them maybe call that figure four is probably the iPhone one moment we're going to have here soon. It's just such a significant step up from where we're at with figure three. But we're really early. So, in the iPhone, even from Gen 1 to Gen 8, 10, every year, I was like, man, this hardware is getting better. Cameras are getting... Now it's a slightly better camera. Now you can't really tell. It's like anatomically approaching some perfect iPhone vision for them. But for us, we're really early. So, I think the robots will continue to get much cheaper, more dexterous, just overall safer to be around. I think a lot of the way you interact with the Humanoid 2 will really improve.

  25. So I suspect if you do a leased model, you can really get the newest version every few years, which will be good. MARK MANDELSKI- It feels like you're next-gen. It's been an 18-month center on next generations. FRANCESC CAMPOY- Yeah, we really want to do our next-gen at significantly high volumes. So we spend a lot of time on it. And we spend about a year between programs right now. We'll spend more than that on figure four. What might we see with Figure Four? What areas are you focusing on making improvements? Oh, man. That's a little... I think the big drivers here at a high level, at the requirement level, are like, we need the robot to get cheaper.

  26. Like, that's super important. We'll induce more demand. So we're trying to reduce costs as much as we possibly can. We're trying to make the robot more capable. You want it to be able to do more things in the world than it has before uh you know like a little sneak peek of that is we've been working on the hands for like uh around almost like two years now and we're like probably the only group that's built like tendon driven hands and non-tendent driven hands we've done now seven iterations of our hand um we i think we really understand hands really well and um one of the biggest step-ups we've ever made in the hands is coming and um i would make the argument that like a figure four with a glove and a human with a glove is almost indistinguishable really are you serious yeah yeah that's like an iphone moment hey the whole the whole thing i've seen so far is like they're functional But even the way they grip things, you know, when it was picking up the cups, it's sort of reaching into the cups rather than grabbing them on the outside.

  27. I hate to say this, but like, because like, I think figure three is probably the best humanoid in the market by far, but it'll make figure three look like a caveman. Really? Yeah. So where are you in terms of? It's funny because the iPhone analogy there is really interesting because, you know, the iPhone three was the first really big hit, but it had no camera. The camera came later and that was a step function change in that trajectory. I'll bet the hand is the equivalent of the camera. You throw that hand on there and it has human level dexterity. I just think like, you know, we've been, if you look at like all three robots, they've been, you know, they look different for us.

  28. We've obviously changed a lot across the three. We've changed like, you know, from one to two, we changed like the architecture of the structures from like, you know, it went fully from endoskeleton to exoskeleton. Like all the loads on figure two, like all the exterior take loads, almost like your aircraft does. Right. And how we build aircraft at Archer. So, it's a very different type of architecture as we designed the robots out. But you'll line all these up, and figure forward, it doesn't even look like the rest. It's just like, what happened here? It was a full deviation from our lineup maturing, and there was just a four-foot fork.

  29. And that's what I mean by, it's so significant of a robot. We had a couple hunches of where things should be headed, and we ended up doing a bunch of work on those. And those ended up painting out really well. So we took a couple of pretty aggressive bets on complexity to get there. Anyway, fair to say, I think figure four would be the largest step up we've ever made between generations. That's pretty extraordinary, because two to three was a huge step up. You've never seen anything like this before. The largest step up is in figure four. Maybe you'll sneak us a peek later after the pod. On the hand, touch is super critical, right?

  30. Being able to get that data.

  31. Brett Adcock · 3 min

    “It needs to be tens of dollars at the most.”

  32. It becomes a very difficult conversation to have across 10 people. Do you think a year from now, if you think about the supply chain coming into this, it's just raw, molten metal and bare wire that you're starting from? If you think about winding your own actuators, like you're literally winding coils, like, okay, but at the end of the day, you're not going to make your own CNC milling machine, and you're not going to make your own die casting, or maybe you do your own die casting, I don't even know, but what does it look like?

  33. You're referring to the conversation at the Gigafactory where raw aluminum was coming in one end and cyber cabs were coming out the other end. I mean, it's buckets of molten metal just pouring, and you're like, you're kidding me. It's right out of a sci-fi movie. I mean, to get costs down and some of those times, yeah, some of that has to really happen. I can just imagine you're here to your team saying, more robots. We want more robots walking around. It's so much fun. It's so much the robots. It's nice though, right? It's amazing. That's not nice. It's amazing. You're living in the future. Even the weekends are all running.

  34. It's a bummer when they're not running. You just come in here. How many do you have at home? How are your kids? dealing with it all? Are they enjoying? You have, what, two or three young kids? I have three kids. Three kids. They're just, like, they're really young, though. They just, like, want to run. They run over and try to, like, jump on the robot. I'm like, you can't really jump on these robots. You know what I mean? We're not there yet. You can't jump on them. And they come to the office, like, every few weeks. They love being here with all the robots and stuff. It's great. Well, so when you walk in now, there's a hundred of them outside, just outside the door. And you don't even feel vaguely like you can't walk through them. You just walk right up to them.

  35. And, you know, they're moving smoothly enough that you don't feel like it's going to slap you in the face. I'm going to have some conversations with them on the way out. Let's talk about the figure face, because that's an interesting, you know, set of choices. You know, the Unconny Valley of making something look too humanoid. Right now you've got a black screen. What's been the trade in your internal conversations on what should the face of figure look like? Yeah. I'm sure there have been a lot of conversations on that. And we still have conversations on this. The face is really tough because we're really missing a lot of the facial movements of a human.

  36. Sure. So when you talk to the robot, it has, like, you know, lights and screen, and it has, like, obviously the gestures it can make with neck and the whole body. But it's just, there's just still something, like, missing to give it, like, a human-like experience where we're, like, we're, you know, like, we have, like, so many features. Do you think you'll evolve the face to something more human-like? Because right now it's a black face. Yeah. It's, you know, it feels friendly, but it's a little bit ominous to look at, you know, a shiny black skull. Yeah, we're working on the face. We have some big plans for figure three on the head and face, and we're working through this.

  37. It's just a hard problem. It's a human psychology problem. Yeah, like, you really want to, like, I think, like, we're seeing more and more internally. We're, like, talking to the robots more. We're spending time with them. The robots, like, have, like, memory now. So, like, if you, like, if, like, we've done work now where you can, like, I can go up to a robot and it's like, hey, Brad, good to see you again. And, like, remember is our last conversation. So you're getting now to a point where you might want to just almost like a companion, you might want to talk to it, especially in the home when it's around family and you've had it for a long time.

  38. So getting this, we call it HMI, but again, it's human-to-machine interaction, really well-dialed. We have a whole team on it. We have an HMI lead. We have a whole focus on it. Every week we meet and do planning on it and demonstrations. It's really hard. A lot of it comes down to not just the face, but how the body reacts, too. We're just pretty expressive with our, obviously, facial face, arms, all this as we're talking. Making small little gestures and nodding. All this really helps, in my mind, building something that's more natural to interact with. So we have a lot of work we're doing there.

  39. And then, yeah, the face is a tough one. I think the thing that we've really stayed away from is putting googly eyes on the robot. I think it's a really... How do I put this? I think this is one of the most complex pieces of technology on the planet. If we're there, putting googly eyes on a robot and saying, oh, it's friendly, and it's safe, it's just a huge disservice. It's a con. It's like the robot has googly eyes. It can literally walk upstairs, open doors. It's one of the most intelligent things we've ever built. So, I think we're trying to hold a certain standard for design that we show what the technology is really capable of.

  40. So, we're working through this problem now. I think it's a really hard one. It's one of those things where we have to get in a room with a lot of people, and we have to really kind of discuss where the future is really headed. And then we have to do a lot of product iteration really quickly. It's something you could test countless iterations of and how kids and adults interact with it. But, I mean, the end goal is to have, on the consumer side, a home companion that you trust and is part of the family. Yes? 100%. You want to be able to talk to a human, and you want to have persistent memory. You might want to spend a more percentage of your time hanging out with a robot than you would having to do work.

  41. In the home, it splits to you really want a companion there in the home that knows everyone, can be really helpful. A kid comes home from school sad, asking why, and then you want something to be able to do work, like we all do. like this morning i was like getting stuff out of the dishwasher like um i can't believe i don't have my robot doing that right now you know what i mean like um but hopefully you know hopefully by in a year like we're doing that pretty repeatedly um so we're like working through you know all those things but like you know we want to be able to connect with real humans every day so on persistent memory you know if you have a pet a dog or something like that passes away you get a new dog you've got a new you've got a new dog but if you have an iphone when you get a new iPhone, you just port everything over.

  42. And the personality has now moved to the new device. So here comes your robot. You've fallen in love with it. It's got persistent memory. You've done a ton of tuning it. I don't like my laundry folded that way, or you shouldn't be doing that. You should be doing this in the morning. All of that state is now in your bond with this device. So does that live with the device, or does it become like your iPhone, where you port the personality to the, and the other one just sort of folds itself up? It definitely does not live with the device. You'll be able to have that pass through any device. And then what if you have two of them?

  43. Do you swap personalities midday? Yeah. We're doing some testing here. We're like, what if you're walking by these hundreds of robots, and they're all talking to you? And they all remember you. Yeah. All the way through. But you're walking by all the way through campus, and you're having a singular conversation. Well, that's always mind-blowing, actually, because when you see them all lined up, and they start doing things in synchronous lockstep, that is just a mind-blow for anyone who sees it. because you think of them as people and then you see them coordinated in a way that people could never be coordinated. So the equivalent when you're walking through is if you're having a conversation and the conversation is moving from robot to robot.

  44. As you're walking down the line, you're like, wait a minute. And that could be good or bad, you know, because like that uncanny moment. The shared mind. Yeah. When I think of these robots, I would want to personalize them, both cognitively but also physically. You know, right now they're all wearing the same outfit and costume. I assume you'll have the ability you can design your robots. It's like your iPhone case. Yeah, outfit. Yeah, you're going to want to design your robot. Yeah. Yeah, for sure. Yeah, so these robots can all be offered with different soft goods and apparel. We have some here.

  45. We can show you on the way out. But, yeah, you'll be able to fully accessorize your robot, give it whatever personality you want, change its voice. What about tools? Are there going to be figure-branded tools that are? What do you mean by tools? Like a hammer? Well, yeah, like a hammer, but, you know, like a mixing machine, you know? I'm making cake, you know? It's like, okay, well. That'd be pretty cool. Like appliances, yeah. appliances like you know i spent half my life in the longevity business the favorite example i use is if i've got a bunch of sensors in my body measuring my protein levels hydration levels peptide levels and so forth my robot in the kitchen should be cooking me a meal specifically for what my body needs at the moment yeah yeah the thing about tools in general though is like right now we think oh i've got all this human-made stuff my humanoid hand controls it we're like well oh wait, now we have for the first time ever the ability to snap on an entire function, just snap it in, or I can mentally control it.

  46. Like I don't need to, if the device has your IQ from Hark or from Helix built into it, I can just mentally control the device. I don't actually have to walk across the room to mix the cake, I can just brain it over, because I can. Turn the oven on without even going to it. Yeah, exactly, it becomes totally integrated in your whole life. Let's talk about safety, because, I mean, one of your goals, you said it in the beginning, for getting this out, is it's got to be safe. Safety, insurance, how do you think about building in the right safeguards for these robots? Because they are very capable. They are powerful to do what they need to do.

  47. What's your philosophy there? What are you building into the system? Yeah. To make it really safe, you have to work from basically all the way down to the substrate level, almost like a non-volatile memory, really understanding what those rules look like and affordances for the robot, all the way up to the system architecture design. A system safety architecture. It's really tough, because you have both physical safety and, I would call, more semantic safety. The intelligence of it needs to also be pretty high. You don't want a robot walking around and knocking over a candle or spilling the hot water. These are things that you need to understand. There's risks in those areas and how to deal with them. Separately, we have 40-plus actuators on board. You don't want any of those failing, you don't want any of the robots falling, especially with, say, kids or pets and stuff in the house. We've spent so much time, say, two or three generations now, thinking about what is the right architecture to build a real-time system that approaches never having faults and issues. You've seen this in the world. We have turbofan engines that are rated to 1x10 to the minus 9 hours for a catastrophic event. You have very complex

  48. systems out in the world that we've gotten to really high safety standards. Self-driving cars, we need to approach this, things like this. I think we see a clear path of how we design a system that gets to a point where it's safe enough to scale into the millions and billions. But beyond the system architecture, there needs to be a lot of iteration design of being able to build and then test, learn, and then rebuild and retest and um and then you basically need to build a track record behind that but it's about like you know these are running software software can get hacked do you have a master off switch that can be you know physically actuated um you know we've seen agents escaping the lab you know it's been it's been all the conversation on the podcast i was with daniela russ who runs c sale at mit yesterday they have a whole bunch of chinese robots that they bought because they're really cool in the lobby.

  49. And they're all transmitting data back to China. They're recording video and audio. So I'm not worried about that for Brett. Well, it's actually an advantage for Brett in the sense that- Can we buy all of our competitors Chinese robots? Having a robot hacked in the home by some outside person. So how do you put the cybersecurity or physical off- command if needed. Is that something you're thinking through? We have a whole cybersecurity team. All they do is deal with this. Both on the product and corporate security side. It's actually a pretty large team as it relates to headcount here internally. Self-driven cars have the same problem. You don't want all the cars in the street to be hacked and driven into pedestrians. Same with us. We have safeguards in place for this work. We're continuing to refine that and make that better. It's an evolving thing.

  50. It's not like we have a silver bullet where we push one line of code, and we're like, yep, that will solve everything, because it won't. We have a tremendous amount of work happening from cybersecurity, system safety engineering, all the way down into, what are the fundamental rules within non-volatile memory that we need to put into the robot so it behaves like properly in the world. Actually, when we met nine months ago, you mentioned you were well down the road to having no Chinese or third party components from outside the country. Have you made a lot of progress on that since then? Yeah, we've made a lot of progress last year or two on this point.

  51. I mean, it happened because like first, like tariffs just got so expensive. Yeah. I think at one point, like last year, we were paying like one week was like 270% tariffs. It was like, you know, during all that negotiation stuff. And I was like, I think we were like, this is going to end soon. That's an incentive. But it was like, yeah, this is not good. And then obviously, like the geopolitical stuff like really flamed. And so, yeah, we've been rotating like fully out of like just different piece parts and stuff we have out of China. So at this point, it's like in a really good shape. Have you seen a robot insurance industry pop up yet?

  52. I'm assuming there will need to be insurance models wrapped around robots. Yeah, it's interesting. We haven't seen any of that. Both commercially and the consumer will need this. In terms of financing the robots, are you going to provide financing? You're going to have outside financers coming in, so instead of buying for $30,000, I have a lease package. We currently have a lot of demand for folks that want to finance robots for us. Right now, a lot of it's focused on commercial contracts they want to finance. There's no reason why this won't happen also in the home. This is a multi-trillion dollar space of doing credit card receivables and lease financing for automotive.

  53. we're working with the biggest companies in the world around this. I don't know if we'll personally finance them, but we'll work with different banks and funds and stuff to do this work. Yeah. Do you think about space at all? Because I know this seems like a stretch because we're in the kitchen right now, but no, there's only like three, maybe four people on the planet that have the wherewithal to train a $3 billion foundation model for physical world. And meanwhile, things are starting to go into space, not just on Elon's rockets, on other rockets too. But if you think about Eric Schmidt launching stuff into space that needs to be maintained or changed or whatever. He doesn't have that foundation model, that physical world foundation model.

  54. So there's probably, you know, these guys in space, but there's also like, no, my model controlling something else. But that physical knowledge model is just like this, this insanely valuable centerpiece. Yeah. It's funny you ask, like, you also mentioned like, you know, like this AI stuff, like really happened really fast. When I started Figure, we called it Figure AI. Our mission was to advance AI for robotics. My whole master plan was, there's going to be convergence of AI and robotics, and it's going to be really powerful. I thought that maybe it would happen 10, 20 years later. It happened next year. Deep Learning started working.

  55. In that, we also talked about a key focus of trying to get the space. It's an area that we're really interested in internally. We've had a couple conversations on this last six months. And one group told us it costs like tens of like low tens of millions of dollars per hour to put a human in space to do work. Yeah. And we're like, wow, that's crazy. Well, it's really bad for PR when things go wrong too. Totally. You know, it's, I think a humanoid should be great at this. Exactly. And that's the whole objection to data centers in space and other things in space is maintenance. I was like, well, wait a minute. Throw a humanoid up there to do it. Yeah, exactly. And how cool is it? Like floating around, you know, no spacesuit. We had a team here a space team here and they're like you know how is like you know are you ip rated to this and this and this for space and i'm like oh my god no like like you know what i mean like uh vacuum rated joints yeah exactly so like i've got a lot of work just well i'll be i'll be candid we are not ready right you you very famously said you're not going to do uh defense work uh with figure is that still the case yeah it is i i respect that i think it's kind of hard i think

  56. You know, I think it'd be difficult to do defense work by also trying to sell robots into your home, doing work in, like, everyday commercial, you know, places that we all know. I think it just becomes really tough for humanoids. I think we need to, like, you need to pick. route. We're, you know, we're picking like, you know, the whole civilian side, which we think is a thousand, hundreds of times, thousands of times bigger than military applications. And it's also like an area where all of us here at the team of the company really want to solve. We want to solve a robot in the home. This episode is brought to you by Blitzy, autonomous software development with infinite code context. Blitzy uses thousands of specialized AI agents that think for hours to understand enterprise-scale code bases with millions of lines of code.

  57. Engineers start every development sprint with the Blitzy platform, bringing in their development requirements. The Blitzy platform provides a plan, then generates and pre-compiles code for each task. Blitzy delivers 80% or more of the development work autonomously, while providing a guide for the final 20% of human development work required to complete the sprint. Enterprises are achieving a 5x engineering velocity increase when incorporating Blitzy as their pre-IDE development tool, pairing it with their coding co-pilot of choice to bring an AI native SDLC into their org. Ready to 5x your engineering velocity? Visit blitzy.com to schedule a demo and start building with Blitzy today. I am curious, you know, when the bird scooters started showing up in my hometown of Santa Monica, people started sort of like trashing them, hanging them from trees and so forth.

  58. So I'm curious, when we start to see robots walking around the streets, how do you keep the robots safe from the humans? Yeah. Yeah, I guess we don't want to be like slapping humans, right? Well, it's more than... That was the opposite question. No, no, it's, you know, there's a significant amount of civil fear, distrust, anger towards technology. And, you know, you saw the Waymos burning in L.A. whenever that was, you know, 18 months ago. I can imagine the robots themselves need to be kept safe from humans. Yeah. I mean, we're certainly not going to train any robots to, like, defend themselves against humans.

  59. And so to the extent that a human wants to trash our robot, I guess there's really not much we can really do. There's a lot of cameras on that robot. I know, but like, yeah, we wouldn't want to like, you know, we wouldn't want to. uh try to defend ourselves or try to um stop us in any way i think um you know i think it just is what it is i think uh i think you're gonna have some of that but i think you know net net it's probably like a like a very small percentage of what really happens you see as way most too but it's you know what i mean it's not like um i i i think it's just somewhere i have to go like basically like build through yeah yeah we haven't had it happen since yeah we're a little worried to like, is there going to be a blue van drive up and like snag one of our guys outside?

  60. Yeah, but again, these things are going to be apart. Yeah, yeah, it's funny. It is really hard to imagine anyone getting away with that more than once. You know, there's so many cameras, it's all connected to the internet. Really funny, the robot's like in the back, like, listen, man, you don't want to do this, you don't want to do this, untie me. Oh, that's interesting. He's like, I have a two kilowatt hour battery pack, I can light on fire and take this whole thing down. Don't test me back here. Brett is not going to, you know, pay a ransom fee. He won't pay. I've already asked him before. Don't do this.

  61. It's their policy. It'd be really funny. I just try and talk him out of it.

  62. Brett Adcock · 4 min

    “It feels for sure within five years where I have like a full Lights Out manufacturing.”

  63. whole closed-loop, lights-out manufacturing that we designed into Figure 4. That's why we're saying iPhone 1 moment. Part of this is we think this is the robot that will build itself.

  64. Let's flip it around and ask the other way. When the new hand comes along, which is imminent, is there anything you can think of in the current manufacturing process where here's a person controlling a milling machine or whatever component, anything you can think of that won't be able to be done by that new hand? there's like blue collar job opportunity what's that last thing that somebody needs not really man i mean i would have said like maybe ladders and this like you know some of this forceful interaction stuff would be hard but like the robots are climbing a ladder next door like it'll just climb over and over every day like rain or shine just going up and down ladders and uh we're like doing like very forceful interaction work now and carrying big boxes around our 50 pounds with the neural network we're we're unloading things in a truck we're doing things in a home, like dishes and folding laundry, like, it's just like all the puzzle pieces are right there.

  65. I mean, the ultimate functionality for me is when is a robot doing surgery, right? When is a humanoid robot the best surgeon on the planet? Well, the timeline from the lights out self-improvement to that moment is so short. I mean, so we've just seen, you know, AI become the best diagnostician. I mean, the numbers are blindingly obvious that, you know, if AI versus human, AI wins every time, right now. And then, you know, the next question is when we have a robotic surgeon so that, you know, it's federated learning. So every robot knows every surgical case ever done. It doesn't have any jitters. It didn't have a fight with its boyfriend or girlfriend the night before. It's there and it's able to do surgery. And what's fascinating is it's, you know, a dollar an hour for the surgery, right?

  66. And any place on the planet. I mean, I think that's one of those. I also feel like I've been like half the time I've ever visited a doctor in my life, I've been misdiagnosed. I feel like just having a network of like the smartest thing in the world with you that can be there with you in a room, checking you out would be like an incredible, I'd be like the world's best doctor for every country in the world. Yeah, we're there now. And then the question is, you know, when are these robots the best surgeons in the world? where, you know, you go and you need something. You say, I do not want that doctor touching me.

  67. I want the robot that does it perfectly every time. I think I want to capture that. So if that hand comes out imminently and we can demonstrate every single step of the manufacturing of the figure four being done with that hand, that's a big moment in human history. And I think we might want to just document that, you know, just like literally every step of this entire product. Here's something that arrives in a box, open, assemble, build. It's like, yeah, I mean, guys, here's what I see. We see basically a task we'll pick out. like, we need to be able to do this or that. We go get data for it. We train Helix. The robot thing goes and does it. And we've done that now over and over now for the last, like, you know, couple years. And we're like, man, this robot's really capable. How do we get it to generalize is the real thing. Like, how do we, we want it to be able to walk into Baku with a minimal set of instructions, like whatever training a human would get, probably some slides and some documents and maybe like shadowing somebody for a few days. We want the humanoid to be able to go that same process. And that's really the chapter that we're in right now, which is how do we scale intelligence? How do we get this thing to do it? From a hardware perspective, the hardware

  68. is going to get as dexterous as hands. Think about it as a steady rate of improvement. So if it has a hard time handling a needle one year, the next year it'll just get better and better at this. This is all now happening really fast. This is like single-digit years. I think that's actually a much higher bar, though. The moment in history that's true lights out manufacturing needs to demonstrate like i hear all the steps of creating a figure four yeah pick anyone at random this robot with this hand can do that step that's a moment in history that is actually sooner and easier than the i gave you an open-ended task that you've never even thought of before and you have to go and figure out how to do it that's actually a much higher bar but it's probably only you know six months i just like the one thing i've learned is like these neural networks like love um love data they love to learn they love the amount of variety of data that's coming in really well like i would have said like um even on this logistics line that we're seeing behind us now it's running the packages are all like you know like moving they're like they're squishy they're moving on the conveyor like the robots in a lot of cases

  69. hasn't seen these type of packages before and um for whatever reason it's just it's just good at learning these representations and training um and uh i would have you know would have thought that's never been the case. We're going out now and doing large-scale data collection. As we're bringing more diverse data, the robot's getting smarter at the things that it's not even... Going out and getting data, some guys I saw has a hose and washing down his dog. Another one is somebody's painting a house. And then we're taking data like this, and it's getting better at folding laundry. And it's crazy.

  70. It's like this data diversity. We've seen this with LLMs in pre-training. This data diversity is just so healthy for the neural net to learn these representations across our world. I don't know if you can quantify this, but how much do you think is general purpose, physical world knowledge, and how much is like, okay, if I want to train you to do this particular type of surgery, then I need exact data from just that surgery. So right now when people automate these things, they think entirely down these narrow swim lanes. But we've never had the general purpose neural net before. Well, we've been on the narrow swim lane for two years.

  71. That's what we do, right? We've been doing that for all. We're narrowly, I would say arguably probably the best in the world at post-training these narrow use cases. What you really want from a humanoid is that it can adapt to things really quickly, things it's never seen before. In the case of, say, Baki, or even surgery, you want very minimal demonstrations of the robot to be able to do this work. That means the prior here, the robot coming into it without any surgery training, should have a decent understanding of, just with a prompt, what it should be doing and how to do it. Then, with some more specifics, I need you to do this task, then go here, then do this, and watch me here. This needs to happen this way.

  72. Without that specifics and context for the robot, it should then be able to do the work. It needs to have a good base understanding of how to do this stuff. Like a human was. A human walks into getting prepared on what their first day of Baki looks like. They're a really capable human. They need a little bit of verbal prompting, or show them how to do things to really understand and latch on to the concept. So for us, it's like, it's really not about these narrow peaks anymore. It's really about like this broad level intelligence that we have, like in LLMs. And then from there, the robots can be fine tuned to a spot where it's like the best in surgery or the best in, you know, like logistics or the best in, you know, some specialized task.

  73. But it should be generalizable enough, though, to do most things. Like the surgery bot coming in should be able to go grab your coffee really quick without any training. You know what I mean?

  74. Brett Adcock · 3 min

    “So, we can do maybe over 40,000 or so units now at full rate out of Baku next door a year, across like four lines.”

  75. And this goes to the point of the conversation you and I had with Elon about UHI, when the robots are actually able to massively demonetize everything, like building your home, doing whatever you need at that dollar an hour price tag, it creates abundance for everybody on the planet, which is a beautiful thing. Yeah, pretty deflationary. It's going to be great. It's going to raise wealth everywhere. We had a late night conversation, and you said something that really was a powerful idea, and I want to come back to this. You said there's sort of an ethical, moral issue where when AI is becoming, when we reach ASI, and ASI needs things to be done, you don't want ASI telling the humans to do these things.

  76. You want ASI telling the robots to do these things. Do you remember that conversation we had? Yeah. Do you want to expand on that? Because I thought that was a really important point. Yeah, I think like, I have like a view that robotics is super necessary to avoid a dystopian future. There's a question right now, are we at AGI, ASI moment here? The digital AI is so damn smart, but it's also not very magical. It's almost like a search engine replacement. It doesn't feel like AGI or ASI, in my mind. It doesn't feel like the Jetsons right now. It just feels like it's fine. It can replace my Google searches.

  77. But as this gets more and more intelligent, the digital systems that are living in a server, in order to move things around to make money and help more in the economy, these digital systems, there's a few... future where they're like bossing around humans to do this work, paying them money to go do the work for them. Yeah. Almost like enslaving humanity. It's... Well, we have a long history of enslaving each other and that is kind of an outcome. There's certainly... There's already some companies popping up that are asking humans to go do work and paying them per hour to let agents go out and get work done.

  78. In my mind, like, if there is no, like, robotic solution here, it can lead them to a really bad place. So my view is that as we build into AGI and ASI, like, the humanoid should be doing this work and not humans. And I think it's a really important part here of the story that it all needs to come together. It needs to come together really fast. And I haven't heard anybody speak about that. And I think it's a really critical point you made that, you know, to avoid that dystopian future. I mean, no human wants to be told by the machine what to do. Yeah, there's another Peter's happening. There's like websites out you can go to me puppets.

  79. Just like it's just like the agent to hire somebody to go open a bank account or go do something for them. A hundred percent or go do a task. Yeah, it's like a like mini like task rabbit type services are up now that are doing this. They're getting real traction. Yeah. There's another type of dystopia that I think we're on a path to avoid because of figure, which is, I don't know if you ever saw any of those Russian cars back in the Soviet era, but everybody in the country has to drive the exact same color piece of shit Soviet car. It's like one form fits everybody. and when you look at robotics in manufacturing today and you look at like a honda accord assembly line it's like this robot is welding it's welding it's welding it's welding and what comes out the end of the factory is an identical honda accord for everybody different color yeah maybe a couple configuration differences but now the llms are capable of mechanical design for the first time which means somebody with an idea for a new device that never a physical device a gutter cleaner a pool cleaner like a companion whatever you can vibe design it but you can't make it so unless all the parts are 3d printed or something like that it'll never come into the

  80. world yeah all of a sudden figure robotics comes along capable of making itself which means it's this far away from being able to make virtually anything so then you're in a world with massively bigger variety of physical things things. Because it's just a completely different supply chain. It's all, you know, anything that can be done can be done very cheaply, fully automated. A hundred percent. Yeah. They're just going to work their butts off. All they're going to do is work. All they're going to do is work. We literally have had that robot since the, like the logistics live stream running like just every day.

  81. It's just still running right now. It just hasn't stopped. Instead of a webcam in a fish tank, you've got webcams on robots doing packages there. Hey, can I ask you about chips? Yeah, of course. So when you look at, you know, the, suppose there are 10 billion robots or more, and suppose figure is a huge fraction of them, maybe a dominant fraction. The compute behind them wants to get smarter and smarter and smarter. So Elon is looking at all the supply chain components, and he's like, TeraFab, this is the one dependency. I can eliminate all the other dependencies. They've got a roadmap. Oh, wait, the chips.

  82. So if we were talking nine months ago like we were, the idea that you would have the capital to chips is about a $16 billion problem. But now you've got billions of dollars of cash. Then suddenly it's viable. Are you going to go down that road? Is it something you're talking about? How far down the rabbit hole are you going to vertically integrate? Yeah, I suspect at some point we'll, yeah, I mean, it's really a question is, like, can we control our destiny well enough with the current supply that will exist over the next five or ten years? Memory, silicon, like, and then, you know, even comes down to, like, the cost per chip and power we're getting.

  83. Even the chips we have now run on robots, they're just still too costly and too high power. If you really want like a really affordable robot, you're going to have to really spend a lot of time in there. Well, also, you know, you remember you were talking about the actuator comes out of the box. It's this big. The people designing chips right now, they are sold out for decades. And so they're not going to start thinking about innovating in the direction that you need to go for robotics. They're just going to try the cheapest, fastest fab they can put up and spew out. And you just have to live with it.

  84. We actually met with the C-suite of one of the largest memory OEMs in the world. And it's like a few months back. And they're like, we have, we have, we, you know, trying to talk them through like, you know, our future and roadmap. stuff like we have no capacity for this yeah all of our capacity is going to like phones and servers everything else and like the phone like even like if a phone like doubles in memory takes the whole capacity of the whole market and the phone and they are like phones are getting choked off you know think about like the amount of margin in a phone is insane and it can't even get the supply turnus was complaining about that his earnings call said memory is destroying us memory is bad yeah we actually um we've done a few pretty cool like memory partnerships here uh, to short memory over the next few years, uh, internally. So we have like, um, uh, in the near term, we feel very confident in our ability to scale up robots. And then, you know, you're asking a lot of good questions about in the longterm. So, um, yeah, we'll see. I think like, uh, as you mentioned, it's a huge capital investment. The, the technologies, um, the models itself and the way we're building the models and running them is really evolving fast. Yeah. So I'd be, you know,

  85. The one concern I have is like building, you know, this chip work takes like two, three plus years. Mm-hmm. So you have to look out three plus years minimum and say like, okay, what is the design going to look like? And if you told me three years ago, what's the design of my inference compute here? It has been completely wrong. Yeah. And then we have like some, you know, work we're doing now with Helix 3, which you guys just saw, that I would say is a radically different type of way we're running now than we're used to. So I think, yeah, it's a little bit hard to, like, lock us into a certain amount of constrained power and cost and, you know, chip architecture, given how things are accelerating so quickly.

  86. So it's been nice to be able to, like, move to the right chip for us now. But, like, but, you know, if we look out over, like, a long enough period of time, we're going to need access to, like, unprecedented amount of silicon. Well, there's a pro tip, you know, there's a gap to, if you think three to four years out, that's the earliest you could get your own chips. But between here and there, Positron, Etched, Grock are getting incredible headway on six nanometer or thereabout process nodes that are still available. But you have to race right now to book that capacity. but a six nanometer node that's directed toward your use case is dramatically better than the two and three nanometer nodes that are you know that's all the iPhones now yeah and all the NVIDIA GPUs are built around the two and three node they can't get any headway out of the sixes yeah but that's the stepping stone that fills the gap yeah but it's you know those conversations are moving really really fast and those companies are you know it's funny in the in the physical world um entrepreneurship world yeah there's there's only like two or maybe three of you so i mean it's a huge bar but someone has to fill that void otherwise we're all going to suffer for lack of robots for for some period how old are you

  87. Brett? 40. 40. So congrats. I think it made it. I think congrats is, you know, you've built serially increasing, more valuable companies, which is which is extraordinary. I think you were making the quote earlier, the stat earlier that what was it? Staggering. It's staggering. So, all entrepreneurs, all time, only Zuck is ahead of you at age 40, but serial entrepreneurs... Zuck? God damn it. It gets better. But you're a nice guy. Serial entrepreneurs, which is basically Jobs, Elon, a handful of others, are far more likely to change the world in the future, which totally makes sense to me having invested in hundreds of entrepreneurs over the year.

  88. If you can do it once or twice, you're going to do it three, four, five times. If you do it once, it might have been a lottery. It might have been just an outlier. It doesn't mean you can do it again. But you've already proven to do it like four times now, I guess. So the forward trajectory of that profile is so much better statistically, not to put too much pressure on you. But then among serial entrepreneurs at age 40, you're 20x ahead of where Elon was or Jobs was in terms of resources that you have available. Let me just phrase it that way. Now, part of that is just sign of the times. there's just a lot more capital available today also we were talking at breakfast this morning about if you i think you were destined to be a physical world guy you know it's just like feels like it's in your kind of growing up in the farm farm dna you know machinery i like touching hardware yeah i do it but if you if you were born 20 years earlier you would have run headlong into the hardware is hard silicon valley is never going to invest in it you know the moment is so perfect right now. So you're just hitting it. It's just going to be so fun to watch where this goes.

  89. Let's talk regulation. So, I mean, so much conversation around regulation of AI, but I haven't seen that in the robotic space. I've seen sort of the U.S. government putting a regulatory block on... on importing Chinese robots, which I get, but are you starting to have conversations or hearing from regulators about, we got to regulate robots, they shouldn't be on the street, we need to be careful about in the homes. Where is that space right now? And what are you expecting? What are you sort of like gearing up for? I think the FCC ban on Chinese robots was like a little bit of a it was a shock um yeah i think from my perspective on that like i'd love to see like a really level playing ground to compete yeah i mean as you think about i'm going to come back to regulation a second but as you think about competition is it is it tesla and china is that the way you see it in the world right now i personally haven't seen it much out a Tesla in two or three years.

  90. I've seen some teleoperated robots, but I haven't seen any. You haven't seen this. Have you? No, I haven't. I've seen China do a lot of dancing, but we've talked about that before. The neural network architecture is completely, it's literally 180 degree different work. If you said, Brett, we need you to outdance China. They do a damn good job. We'd have to pivot everybody's strategy. We are dancing roadmap and go dance. It is literally 180-degree opposite work. We need to get a bunch of humans wearing Hollywood suits, dancing around. We need to then work on replaying that on robot. And we need to get out in the middle of the woods and go around and leap over rivers.

  91. Yeah, so it was always Boston Dynamics was making the best robot dance videos. I mean, listen, I think if I answered your question, Even more directly, I think if I had to pick the closest competitor to figure, I would say it's certainly China today, like the collective China. There's a lot of work happening there. They're moving quite fast. I happen to view they're moving a little bit different direction than we are. And there are a lot of them. There's a lot of them. Yeah, there's a lot. There's like 150 humanoid robot companies. Yeah, it's kind of all crap right now. And the hardware is not great. The policies are running and models don't look that great. But the amount of progress they're making year to year is really great. So, I would expect China to continue to do much better.

  92. It's a group we intend to beat. Sure. I mean, China needs robots desperately. Yeah, really bad. For manufacturing and for elder care. We need, too. Yeah, of course. But are there any of the particular Chinese robot companies that you think are doing a better job than others? Not really. I think we care about something very different than what we think China cares about. China's almost in this race to show how capable their robots are in gymnastics. We care about, like, how do we ship a robot in a home to do, like, zero-shot, long-horizon reasoning. And you're not seeing that out China. You can't even, like, these groups, like, even, like, Unitree, they don't build good hardware when you, like, buy it.

  93. They're, like, super tiny. They come out of the box. They wave. With the controller, you're like, what else do I do with it? It's unclear. They have no hands. And they have compute the size of, like, a USB stick. So, you basically can't run any real, like, models on them. Seriously, like, all the groups that are doing research put, like, compute backpacks on them. And the sensors aren't great. The hardware was basically duct taped together. It wasn't manufactured well. But will they build better robots over time? Certainly. They're going to build great. I think over time, probably great robots.

  94. We just got to go beat them at this. It's a real race here. So we're spending all of our time and energy thinking about where do we want to head long term? And we're trying to move that left on the time curve as hard as we possibly can. And every morning I come in thinking about how to move faster. Because there's always a way to move faster. And for our entrepreneurs out there, hear this. There is always a way to move faster. And speed and agility is your greatest. You can literally always move faster, which is like quite an unbelievable thing to have access to. You can even just, you know, working harder, like work double the amount of hours.

  95. You will probably make an order of like one and a half, maybe even two times more speed per year. So like all of that really matters as we're competing against like the, you know, the global collective of China. Welcome to the health section of Moonshots brought to you by Fountain Life. The discussion in this section is for informational purposes only and is not medical advice. Please consult a qualified health care professional about your individual needs. I'm a co-founder of and client at Fountain Life. You know, AI is having an outsized impact on every aspect of our lives, how we teach our kids, how we run our companies.

  96. It also is having a huge impact on health. helping you prevent heart disease. One of the key things. I'm here with Dr. Dawn Musalem, our chief. medical officer at Fountain. Heart disease has been personal for you as well, hasn't it? It really has, Peter. And my daughter was five. My husband died of sudden cardiac death. And so this is a topic that is one that I am mission-driven to try to eradicate. Prevention first and early detection is absolutely critical. 50% of people die of heart attacks with no warning signs. No shortness of breath, no pain, no nothing. No, silent killer. They just don't wake up in the morning. They don't wake up. And so, you know, AI, this is our mission to advance science, to try to help to one day democratize wellness. We know at Fountain Life, when we do this CT angiography with AI analytics, we're actually finding that 88% of people coming in have detectable coronary disease. But Peter, what's more alarming to me is 23% of those individuals had soft plaque. This is the plaque that would not traditionally be seen on CT looking at calcium scores alone. And this is the plaque that we must intervene with, with the multimodal testing we're doing, including diagnostic laboratory studies partnered with healthy lifestyle recommendations.

  97. So listen, make sure you understand what's going on inside your body, genetically, metabolically, and cardiovascularly. You can know, and it's your obligation to know. So check it out at fountainlife.com slash Peter to find out more and really make sure that you're the CEO of your own health. All right, back to the episode. You have the advantage of being a founder-led company versus a large industrial company trying to get into the space. You're willing to take bigger risks than almost anybody else. You're looking at disruptive strategies and approaches and iterating rapidly. When you walk around the building here, the talent level, you'll never find it at a big industrial company.

  98. I mean, yeah, there's like a bunch of like small little things here. Like, I think the only way to really build like really capable general purpose robots is in the Bay Area. Like we invented almost like robotics. That's better than Shenzhen. We built robotics here. It was AI was literally invented here. You have Silicon was built here. Like everything, you know, have batteries and motors and embedded software and thermals and AI, deep learning, all of Infra, all of it was built here. You're talking about the best place in the world to go do this. Self-driving cars were here before they were here in China.

  99. AI was here before it was in China. The internet. These are all things that were birthed here. And the people growing up here have worked in these areas and really understand the technology as well. And we've assembled what I think is a top tier of them. And the whole team is a bunch of maniacs. They're just like, they're really great at what they do. They love being here and mission-driven. And we're all here to kind of beat everybody in the space. What was your journey to Silicon Valley? Because you grew up in Illinois in farm country, right? And then you went to college in Florida? Yeah, I went to college in Florida. I moved to New York City after that. And then I started a software company called Vettery and ended up selling that in 2017 and still in New York. It was in New York. We, it was basically an online, like recruiting marketplace. And so we had like a, basically a two sided marketplace to bring on employers and job seekers to help with recruiting and ended up selling that the world's largest recruiting company in 2017 for a little over a hundred million. And, and then as soon as I got the term sheet for this, I was like, I need to, I had a few things I really wanted to do in the world and a

  100. A few of them were in the hardware space. And then I ended up starting Archer Aviation like that next year. And then- Started it here or in New York? I started in New York with the intention to move. And it was first I was going to move to LA because like all the, like, you know, some of the best aerospace folks in the world. Yeah. And then I found out it wasn't really what I was working on with Archer, like building like electric, like, you know, like basically like electric, like electric, it's basically like robots it's like electric motors and battery systems um and control software and stuff none of that was like stuff that aerospace was building um and it was all happened in the bay area uh you had some groups out here building like like e-vital aircraft like kitty hawk and airbus and a few others and you had like all this electrification community yeah and robotics community and so i moved here uh moved here then and i've been out here now ever since um and it's been it's been great. I mean, it's like, you know, Archer is basically like a flying robot. And it's got all the same, you know, it's got electric motors, batteries, control systems embedded software, sensors, like it's a flying robot. And so, you know, doing robotics out here

  101. is arguably the best place in the world. Maybe second to like, I think, number one, it's got to be the Bay Area. Maybe number two is maybe Shenzhen. Those are the places that you'll see, if general robotics will be solved, you'll see it in one of these two places. I totally agree. It's amazing, actually, if you'd stayed in New York or moved to LA today, you know, with $9 trillion of new liquidity from Anthropics IPO and NVIDIA's run up and everything, if you woke up tomorrow morning and said, you know, we were on this podcast yesterday, I think. We need to start building our own ship, Fab. I need $20 billion.

  102. You can do that here. Cannot even come close to doing it in New York or LA. This is the new capital in all those ways. You know, I was in New York for a long time. It's like, I love New York. New York's one of the best places I've ever lived. And I was like, man, what's this deal with SF? Like, I'm in New York City. They had, you know, probably like in software wise, probably the number two market in the US besides SF for like doing software engineering. And I was like, I don't really get this like SF thing. and, you know, end up moving out here for Archer, and man, nope, do I get it now. It's like this most condensed place in the world of talent that are all kind of in this right cultural phase of their life to come over here and really, really, really punch it.

  103. It's actually the easiest place to import talent. Like, if you go back to Chicago or Boston, there's a huge amount of talent. You can't get them. But you can migrate anybody to here. To the Mecca. If you're in like Austin or Chicago, whatever, trying to do robotics or Boston, you'll get some folks, you know, like one out of 50 or so, two out of 50 to move to move there from SF. But like, and that's if you get 50 out of 50. Yeah. And it's in like in robotics, like there's like an order of 12 different divisions of folks you need to build one robot. So it's like, OK, good luck getting those 12 people to come in those 12 groups to build out by hiring them all from SF.

  104. It's just not going to happen. It's just too difficult. When you moved out, everything you said was true when you moved out, except it wasn't the financial capital of the world yet. Now, with all the new liquidity, it's all that plus it's the financial... Like Sand Hill Road and the SS. Speaking of that, you took Archer public, what, $2.7 billion or thereabouts. Peggy Johnson just took Agility public. When's the right time for a company like Figur to go public? What do you need in your mind, having gone through a public offering? Well, I think Peggy went public because she doesn't have money. Yeah, she needed capital.

  105. Yeah, of course. No, I understand that. I'm not going to peg you for years, and I'm not going to speak about agility, but what are the criteria, having learned what you did with Archer? Yeah. I mean, listen, a high level Archer went public because we needed the money. We were raising our Series B, and we needed a few billion dollars to get through FA certification, which is very difficult to go do. I was just not seeing it in the private markets. and ended up going public and raising a little under $1 billion. It was the right move at the time, but we were really early. We were just a really young company. We were too young at that stage to be public, but I needed the capital, so I took the company public.

  106. As a choice, I would have done it over again, just given the constraints I had at the time. But if we had access to the private markets, Archer would still be private. We'd have been healthiest for the business if we were even today. So, on the figure side, I think it's somewhat inevitable that we'll be public at some point. Access to the right public capital. Public trust, as well, for disclosures and such. Totally. I think it's great for employees and liquidity and everything else, for folks and investors. And full disclosure for the audience here, I am an investor in figure. Yeah, early investor, too.

  107. Early investor, yeah. Very proud of it. Before we were cool. You know who I invested? It was you. It was specifically you and the team you were able to assemble. Yeah. And, you know, again, just a note for the entrepreneurs listening, the signal was so strong based upon what you had done as a serial entrepreneur, but even stronger, you were putting your own capital into it. And the team that you had assembled, it was like, that's a rock star team. Team was lit. Still are. Still lit. I couldn't have recreated it again if I tried. It was just right place, right time, and team just unbelievable. And, yeah, it's been great.

  108. You know, listen, I think one of the big focuses for us right now is we want to solve what we call general robotics. This is like a general purpose robot. And he saw a little glimpse of what that looks like today, a little future. And I think it'll get much, much better. That will really kick off significant-scale manufacturing, even maybe the larger significance in inference and compute that we might need as we scale, and just overall working capital and capex needed to grow the business. So, I think we really want to hit that milestone privately, and we feel like we're approaching this. It feels like, man, every new model we train, we're like, is it the one?

  109. And then there's bugs we want to work on and all this stuff. I think we have, you know, sitting in our servers today, 10 times more data that we can train a new model on than what you guys saw today. Wow. Really? I think it's a little... We have 100x to compute coming on. Sorry, I wish I should make it bigger. I think it's maybe way more than, like much more than 10. Maybe a factor of like 15 or so more data. Data sitting there. I mean, we bring data in. Is this data captured from humans wearing cameras? Yeah, it's from a variety of sources. But basically, we need really rich information about physical world information that we can't get from online and scrape online.

  110. You just can't get it from the Internet. And we need to know how would a human grab this cup and how are they orientated and what are the forces they feel? What are the hands doing and joints? The sources, the feedback. The video is all over the internet, but the feeling in that feedback... Yeah, totally important. Yeah, can't get it. We basically have an effort we call index here, which is a large-scale data collection effort. But when we get data in, what I'm trying to say is, we have to go audit that data. We have to make sure it's on the right distribution. We have to make sure we clean it, annotate it, get it ready for training. That's something that's time.

  111. And then when it's ready for training, we've got to be at the right place where we're ready for a new checkpoint for pre-training and go through, it's not like we can just instantly train on it. We have like, you know, most of our compute,

  112. Brett Adcock · 5 min

    “I was like, in 2027, you're not going to be able to compute anymore. And that happened in, like, September of 2026 for us.”

  113. And I would love to ask you, what's the relationship between Hark and Figure in terms of where is Hark going to be powering part of the Figure ecosystem? Yeah. Well, maybe I'll give you a quick overview of both just real quick, and then what we're doing. I mean, Hark's on campus here, 100 feet away. It's nice. You have your children next to each other. It's nice. I don't have to travel and I can always be around and help wherever I can. Hark's basically trying to build personalized intelligence, so almost like the digital version of what we're trying to do at Figure.

  114. We do a lot with the computer use and speech and memory and vision and a few other things, and we're building our own AI devices internally, which I'm really excited about. out. We launched Hark Pro, which is our digital version of Hark, our non-hardware version. We're about a month in now. It's been, honestly, a crazy launch. We were really interested in hardware. We think hardware is going to 10X this whole thing. I was even debating whether we want to put out anything before hardware. They were like, OK, let's put some stuff out. We think the models are pretty good that we're developing now. I actually use them all the time. That's all I've been using for a while now. And I was like, let's put them out and see how it goes. And it was just crazy. We, um, we, we've now had over like 1.2 million signups. Uh, we've done, you know, um, we do, you know, probably like, I think almost like 5 million browser sessions, like since launch, like, um, the usage is just like going through the roof. Um, so much that we had to like curb some countries like last week, just cause we're, Yeah. Anyway, that's good. We're developing models and hardware. Right now, we're very focused on one of the areas that's in real-time speech-to-speech. When you talk to the robots

  115. on the way out, they're using the HARC Voice real-time speech-to-speech model. We're doing a lot of improvements there. I think real-time speech-to-speech is going to be... a really important part of the UI layer for robotics. But there's no third-party components, just totally internally developed, real-time speech-to-speech. Our speech-to-speech model is our own model here. There's a lot of people demoing speech-to-speech on top of something else. It's very similar to what we're talking about with the Chinese robots. The difference is night and day if you have a vertically integrated... Can I tell you something about speech-to-speech?

  116. Speech-to-speech today kind of sucks in the world. The hardware is not the right hardware to be using it on. We don't use it as a primary thing. It's like a thing we'll put on every once in a while, on the car to talk to. It's not really that human-like. It doesn't do well with tool calling, like go and get information. It's not very good, like low latency. You interrupt it, sometimes it doesn't hear you. It's just not quite great. So we, at both Figure and Hark, strongly believe that speech is the right UI layer, the default layer, into robotics and AI hardware. So we're all in on speech. We really want to make that work.

  117. So we're working on that at Hark. We have a collaboration with Figure on this. Figure was the first beta user of Hark. We've been using Hark now for five months. We don't force people to use it, but if they want to use this, they can use Hark internally. The numbers are basically through the roof, internally here, which has been great. We can help with productivity around the company. We're really big at Hark on larger scale, multimodal pre-training. We think it's really important for the space. Obviously, Figure has been working on how do we generate action predictions out of the robot. Anyway, we're doing a lot of... There are some areas of classic collaboration we're doing.

  118. We have a running bet inside the Moonshot Mate team here, especially with Alex. Elon very famously started XAI, then SpaceX bought it, and then X and so forth, consolidating the companies. So the bet is that figure will buy Hark eventually. I mean, I would probably make a different, like, yeah, I would say like, this is just my guess, so don't hold me to it, but I bet Hark is bigger than figure in sub six months. In sub six months? Maybe sub 90 days. Are you serious? From a revenue standpoint? I mean, it's already bigger than figure from revenue standpoint. We've been growing like crazy. And like the Hark numbers are just through the roof. We have some some very big stuff coming on hardware and some very big commercial agreements.

  119. Yeah, I would say, like, you know, given that it's digital, we can scale very different, like, we're in a different slope than scaling robotics. But I think, listen, this is my guess. We don't have anything going on right now. But I think, you know, I do see a path where, yeah, certainly, I think I do see a path where in the next six months, there could be a chance that the Hark's bigger than figure. And then at some point, longer term, maybe figure gets bigger again than Hark. I don't know. A trillion here, a trillion there. The reason I wanted to work on Hark is I was sitting here doing the human-owned robotics thing.

  120. We're an AI business. We want to build the AI brain that fits in a ... The hardware is just the distribution for the AI. It's like the vector it's going to head out into the world in. I was playing around with ChatGPT and the rest of the AI stuff, and they were just so stupid. I was like, I'm just talking to this chatbot that was a search engine replacement from Google. And I was like, this isn't what I really want. I want something that can be with me everywhere I go, that knows me really well, has access to my accounts, and can do things for me, like pay the property tax bill, remind me of something before I forget it, help book my travel for me, do something more complicated that might be long horizon for me, with my permission.

  121. And almost like Jarvis would have. And nobody was doing it. And OpenAI led my Series B at figure. And I, you know, got the vibe of being close to those folks that like they had no real interest in making this a huge priority internally. And I also feel that on the consumer side, like interacting with, you know, say computers and phones with AI is just like a really broken experience today. So I sat here and said, okay, um, I'm kind of in my prime as a builder. Uh, I I'm trying to understand AI more than more. I'm like fascinated by this digital version of AI. It's like even going to a computer and using it.

  122. It's a little mini-humanoid. Like moving a mouse and keyboard and trying to get everything to work. I'm just fascinated by it, personally. So, I spend a lot of time researching it, understanding how basically deep learning works in this space. So, I got to a point where I'm like, I'm just going to sit here and watch all this stuff. Or I can go build and come off the sidelines. I decided it would make me a better founder and CEO if I came in and helped build something that I think was really missing for the world. I think it's going to be really special. There's going to be a point in time, you know all those little charts you see over time?

  123. span of like your whole lifetime, like how often you see your kids and your parents and everything like kind of goes up, then it goes down, like it goes down, down, down until you're like, you know, the rest of your life. I think there'll be a chart here of like both in robotics, but certainly for like, even for stuff like Hark, where this line will be going up all the way to the end. And, um, it'll be such an important, you have a smartest person in the world with you everywhere you go. That's, that's, that's, that has the same interests as you do in helping you. And it's trustworthy though. So the key is that that's actually common between Hark and Figure is that you have to trust this entity with your most personal information for it to be effective.

  124. We tell the team every day, it's like you have the person's overtime diary and we need a new level of security, a new level of privacy that's never been done before. Like a thing that we do, you know, at Hark is we try to protect that at like every angle and it makes it really hard on us. It's like, we can't do what other companies do as it relates to advertising other stuff for their products, because we're trying to protect the interests of the consumer as much as possible. So, anyway, I think... Well, congratulations on a meteoric growth. I mean, it's been... When did you announce Hark? It's not...

  125. We're like over a year old now. Is it a year? We launched Pro a month ago, so we're like a month out of Pro. It's been great, man. I'm learning every day. We're in the early, early chapters of this AI, almost like in 1999, back at the internet. It wasn't quite the right internet yet, but it was all getting seeded. It's all getting seeded now. I thought for a minute, the last six months, we were real late at Hark. I've now realized we're pretty early. We're launching, I think, four or five pretty big product announcements this week. We launched one this morning. There'll There'll be a lot, you know, and then we think about as we like think about getting into hardware, this thing is good.

  126. We have hardware now at Hark in the lab. And dude, is it crazy? So peripherals? When you say hardware, is there going to be, you know, sort of Hark glasses or? Can I give you my thesis here for AI devices? So my view is that there's like there's like two different like areas for AI devices. One is, there's this centerpiece for AI devices, which I think, for today, for consumer electronics, that sits the phone and computer. Call it a billion units a year devices. So, it's right there in the middle. And then, those are the main platforms. And then, around there, you have accessory devices. This could be like...

  127. AirPods, or watches, or glasses, or whatever else. I think AirPods is 3% of Apple revenues. So, Apple's not sitting there getting super excited about the AirPods, but it's an important platform play. They're super interested in the computer and phone. That's the main thing. What will happen is there will be AI devices that will kill those in the middle. They will do a billion units a year devices that will be way better than your phone and computer today, and then they'll be, just as similar as it is now, accessory devices on the outside. Those devices will be single-digit revenue percentage, and they will only work with the main show.

  128. For AI devices, we've only seen the accessories today, like the metaglasses, which are horrible, by the way, and other types of things on the periphery that are not the main show, that don't work well today because they interface with things that are in your main platform. Your iPhone fundamentally will never interface well with the AI device. If you try the Meta glasses, it has to have your app open at Meta on your phone, or it can't get a network. Even with that, it's a horrible experience. The products may be one of the worst products I've ever tried. How do you really feel about this? I mean, let's just get real.

  129. I feel like when I watch the spectacles, I feel like I'm taking crazy pills. Nobody's going to wear these. They keep getting bigger and uglier somehow in this space. But like, you know, so for us, you have to go build it well. So what's a form factor look like for the ideal AI peripheral device? We're not going to tell you. Come on, we're telling you, buddy. Like, we're going to tell you. Like, I think my analogy, my best analogy. But you have a vision for what it becomes. Oh, we have the hardware in the lab right now. Nice. Oh, wow, really? Yeah, we have it right now. It's in the lab. Right next door.

  130. Right next door. I'm using it every day. Does it have windows? Can I look? Unbelievable. It's unbelievable, guys. Like, I, it is just, this stuff is going to be really incredible. Are you coordinating with Johnny Ive and OpenAI? You guys, like, talk to each other? No, we're not. We're not coordinating. Like, not at all. Not at all. We're not talking at all. I mean, we, like, we fired them, like, two years ago after the Series B work. And it just wasn't a good fit for us. But how so? Is there a backstory there? I mean, they, so they came in, I met Sam, you know, and they came in and led our Series B figure.

  131. And they co-led it with Microsoft. And they ended up joining the board, and we did a lot of work together internally. We worked with them for about a year or so. And it was like, you know, the goal was to build like... intelligence in a robot. I was in every weekly meeting. We had a whole team dedicated to it. They had a team dedicated to it. We're like, great, let's get the figure every day, and let's rock. They're like, yep. And then they never showed up at the office. They were like, it's too far from SF. It's just loud over there. We got grease on us, and it's loud. We're working on robots. They would never show up.

  132. We couldn't get them there. I was like, OK, let's do a big event. We just couldn't get them to the office. And then every week I was just watching the progress that we were making, and then we were sharing with them, and then they were making and sharing with us. And it was just completely lopsided. We were making 10 to 1, 100 to 1 progress for what they were doing. And we were sharing all the stuff we were doing. Meanwhile, they're sharing with Microsoft. And then it became pretty obvious. I think after seeing the results from us, they really wanted to get into robotics again and then it was just like in you know it was just something i just when did johnny i've come on board at open ai during this time was that after i don't think johnny's working on the humanoid uh open ai i think he's working i think he's my guess is he's working on the like ai devices work yeah um i don't know if he's what's the iphone killer but now they got afraid of saying that but it's it's some device that has ai embedded in it that's so compelling but your point about there being a centerpiece your personal information isn't going to be everywhere you're only going to trust a limited number of center centerpieces and so i

  133. think they want that device to be the centerpiece not the iphone you obviously have the same vision with hark yeah so i can see how it's listen it's certainly not going to be the iphone like it's uh the iphone is going to go away like all right we're calling it here the decline of apple i mean like like let's just call like and you like listen if you take the stream case in 10 years we're not using a traditional phone yeah you know what i mean like i just like uh like so we know that. It's like Ternus really has a challenge right now because they've been, they've been gliding forever on the iPhone and it's going to get challenged from a dozen different places.

  134. Man, I've like, I've like had this thing where I've like, I've like thought I'd wake up at some point every day and see, you know, Tim Cook or whatever, John, get on some like emergency, like, you know, like PR events and basically say like, Hey, we like, you know, we bought a hundred billion dollars of gpus we're bringing the top team on the world we're going to go at pre-training hard in the meantime we're going to use other models and this is going to be so core to the experience interface with consumers long term that we really need to own it understand it it and embed it. And it's been crazy to see nothing. And it's now crossed the point, probably about a year ago, to a point where it's like really too hard to turn around.

  135. You think? It's for sure. It's like this... It has lost so much momentum. Yeah. You want to co-design the hardware and the models together. You really want to think about these AI devices as just a pure interface. It could be BCI in the brain. It could be language devices that we're working through now, like, this is just the interface to this raw intelligence in their side that's going to be human-like just for you. And they need to be co-designed together. And it's not something you just like flop in. And so instead of like... Are you interested in BCI? Yeah, I think like, I'm very interested in BCI. I think it's one of the most... I think it's a very important vector to be on for kind of like AI interfacing long term. I just...

  136. I'm invested in four of the companies right now. And I've just seen some, I just had some meetings yesterday. It's blowing my mind. We're on the precipice of being able to actually get to a mouse connectome, and then a human connectome, and then the world changes. Yeah. That's sick. We should debrief at some point. I would love to make some introductions. It's an amazing company. Yeah, I think we're in this phase of like, kind of beyond where we're at now. It's like language device phase. I think there's a phase kind of after that of BCI and alignment into the brain. I think it's going to be really important.

  137. I expect Hark to be working through that whole process all the way through. So I think we're like... But certainly right now there's like an upgrade cycle coming for traditional computers and phones. And what I can tell you by just using some of the products here internally is it's going to be phenomenal. It's going to be a much better experience. We haven't changed in 20 years. It's literally 20 years. and we now are just plopping AI into it. And it's like, it's kind of like the interfaces are just really reactionary. Like you have to kind of control the phone yourself, but we've, man, we've built this thing that can think now and it can use computers and it can speak to you and it can see and it can build persistent, almost perfect memory.

  138. And, you know, you want to be able to interface with that real time and even, you know, and have it with you everywhere you go. And we don't have anything near like that in the world today, but we will shortly. Like it's coming. Awesome. Yeah. well after ai is the interface you basically use the phone as a camera but the cameras are like three bucks it's like like okay what why did i pay 1500 bucks i got a I mean, phones are incredible. It's just like there is just like a next leg upcoming. It's going to be great. Yeah. It's about time. All right, let's wrap with the conversation on regulation, because I am curious what you're thinking here.

  139. We've got Bernie Sanders and et al saying, you know, we got to stop, slow down, prevent RSI. And I have to imagine, I don't know if you're in those conversations at all, but I have to imagine there's going to be a similar kind of like slow down. We don't need robots on our streets. It's not safe for the citizenry. How do you think about that? Are you having those conversations? And is there any, you know, is there real concern that the public should have in this regard? I can speak for our plans. Our plans are to launch a really high quality product that works really well into the market. And it's safe.

  140. And it's safe. Like, we don't want to be out there, like, you know, tele-hopping the robot around and doing unsafe work, shipping too early. So we're making incremental steps now. We shipped first commercially last year. We're shipping even more this year. Have you seen any inkling from regulators to start speaking about humanoid robotics the same way they're speaking about superintelligence? We really haven't seen that, like, public-facing. We've seen the only thing we've been aware of the last 90 days is the FCC ban on Chinese humanoids. Beyond that, we haven't seen anything come across our desk. Also, a little bit personal to me, I don't fly to DC every week and try to get my nose in this. I'm here trying to build shit, and I want to build awesome robots. Regardless, there's going to be robots coming, they're going to be great for the world. I think the same with AI, it's going to be great.

  141. If humanity woke up one day and just didn't want AI, we can just shut down the internet. We can just close it. But we don't want that. It's going to be really helpful for the whole entire world. And there will be whatever, Bernie Sanders of the world or whatever here. But for us, we're just really focused on the mission of trying to build and ship your robots. And we're here every day trying to go do that. I got a question just to inspire all the entrepreneurs out there. So I grew up a big Steve Jobs fan, too. and i think a lot about how much more he could have done in his life but in that era you could only do one thing at a time and nobody had super voting control and all that. And then when Google came along and then Zuckerberg came along, it switched to this era where the founders control the companies and they could be much more aggressive with the business plan. And that works really well here in the Valley. But then Elon came along and reinvented it again, where you can do multiple things at the same time. So this is working out incredibly well for Figure and Hark. If you just said Hark nine months from today could be bigger than Figure, it was just a mind-blowing statement. I think you said three months from today.

  142. months from today i mean like harks is going like this guys like i just like it's it is going really well like so um we'll see you've also got this other thing about school safety you know that is yeah which is you know not going to be bigger than the other two companies but it's important for the world yeah happening concurrent like the power of an entrepreneur today is so much bigger than it was in the steve jobs era let alone the zuckerberg era so i think it's really inspiring do you think a 20 year old entrepreneur today can walk in your footsteps or is ai going to make it all too late and it's it's you know the ship has sailed yeah i mean i think one thing i'll say is i'm like completely dying like three kids like three companies like i'm just like i'm just i'm just like i've been here every night for a whole year i'm just like um man has it been hard um but it's like you know it's the best time in my life i never want to do again you know it's like uh yeah it's been it's been going great uh yeah listen i like the future generation is going to be great. Everything's going to be AI-native. You'll be able to build any type of hardware just through...

  143. Prompt the product. Exactly. Just through talking and robots. It's going to be so cool. Really, where the AI is really missing is, it has no imagination today. We're going to be bound by this imagination. This is what entrepreneurs do best, right? They always have a little notepad. I still have a notepad of all the business ideas I can think of. I still keep them to this day. You're a great idea generation, great about thinking about the future. This is going to be an unbounded opportunity for every new entrepreneur behind me growing up. It'll mark the best time in the world for entrepreneurs in history.

  144. Now you have an ecosystem here today for entrepreneurship. When I was graduating school, it wasn't conceivable you'd start as an entrepreneur. Like, it was just like, you don't do, nobody does that. Like, uh, so I had a job for a few years before I started it. I wish I started even sooner, uh, starting companies. So, um, but like, you know, now. You can go do that. You can get capital for it. There's incubators. There's people who want to work at startups. It's fun. You have this whole community built around it, too, which is just going to propel these future entrepreneurs further. I like to joke we've all been promised a three- and four-day workweek. We've invented a 10-day workweek.

  145. I like working, man. Let's just call it what it is. This is the most exciting time ever to be alive. I like grinding it out and working this. I think everybody here with me likes it, too. We have a bunch of groups that like to work hard. And we've made it this dirty word out here in the Valley. So, a lot of folks come in here and be like, you guys are the most un-Silken Valley company I've ever seen. We work hard. We come in on the weekends. We work nights. We just like it. And there's not enough of that anymore. It's just almost like we interview people like, oh, what's your day-off strategy and work from home?

  146. We don't have any of that shit. You come in here every fucking day, and you work with us, and we're going to work with you. You know, my Baku guys can't work from home. They've got to get on the line every day. And same with us. We've got to get on the line too. So anyway, it's just – What you're building is so much cooler than virtually any day job. I mean, it's got to be – even if you're working twice as hard, it's got to be so much more rewarding at the end of the day to look at what you've physically built. That's just – also, I didn't anticipate how much the office environment is elevated by the walking around.

  147. So much better. It's bizarre. It's like it just adds energy to the building somehow. So a year from now, we're going to come back and do our figure four recap. You come back, figure four, figure four gets you all set up, checks you in, gets you water, coffee, walks you around, gives you a tour, and then I'll see you for the interview. Yeah, we'll include figure four in the interview too. Maybe just like honestly, figure four can speak on my behalf. That's so fun. It's like, you know what I mean? Buddy, thank you so much. Thanks for having me. Yeah. Thank you. And now when we cut cameras, we'll go see your secret device in the figure four.

More interviews