It feels like you can’t scroll through your feed without seeing a new video of a humanoid robot doing something that seemed impossible just a few years ago. And right in the middle of all this hype and money is Figure AI, a startup that’s pulled in jaw-dropping investments from tech giants like NVIDIA, Microsoft, and Jeff Bezos.
Figure AI has a massive goal: to solve labor shortages by building general-purpose robots that can work in our factories, warehouses, and maybe even our homes one day. But is the company really on the brink of changing how we work, or is its dizzying valuation built more on a cool promise than on what’s actually possible today? Let’s get into what Figure AI is, how its robots work, why investors are so excited, and the serious hurdles it still has to overcome.
What is Figure AI?
Figure AI is an AI robotics company that popped up in 2022. It was founded by Brett Adcock, an entrepreneur who has a knack for launching ambitious companies like Vettery and Archer Aviation. His mission with Figure is to build a humanoid robot that can do all sorts of jobs safely right alongside people.
The big idea is pretty simple when you think about it: our world is designed for humans. Instead of making businesses completely redesign their warehouses or factories for specialized robots, Figure wants to create a robot that can already walk up our stairs, open our doors, and use our tools. They’re aiming to put these robots into the workforce, starting with industries like manufacturing and logistics where it’s tough to find enough workers. It’s a bold picture of the future, one where robots take on the dangerous or tedious jobs, leaving humans with more creative and interesting things to do.
The Figure AI technology: Ambition meets artificial intelligence
Figure AI’s approach is a cool mix of advanced hardware and smart AI. To really get what makes these robots go, you have to look at both the "body" and the "brain."
Why a human form factor?
The choice to build a robot with two legs, two arms, and five-fingered hands is totally strategic. Like we said, our world is built for us. A humanoid robot doesn’t need a special ramp to get over a curb or a weird attachment to turn a doorknob. In theory, it can slide into existing human workflows without causing a huge fuss. This could give it a leg up on all those wheeled or stationary robots that force companies to make expensive changes to their buildings.
An in-house approach to embodied AI
At first, Figure AI teamed up with OpenAI, using its powerful language models to give its robots a brain. But in a pretty big shift, the company is now building its own AI from the ground up. CEO Brett Adcock has talked about the need to "vertically integrate robot AI," which is a fancy way of saying they need to control the entire process to solve the unique problems that come with getting software commands to translate into smooth physical actions.
The brain of the operation is the company’s own neural network, called Helix. It’s a vision-language-action (VLA) model, which means it’s designed to see the world through its cameras, understand spoken commands, and then figure out how to perform complex tasks. Instead of being programmed for just one job, Helix is meant to learn by watching, which should allow the robot to adapt and handle a bunch of different duties.
This video introduces the Figure 02, showcasing the advanced capabilities of the humanoid robot from Figure AI.
Market hype: Unpacking the valuation and partnerships
While the tech is impressive, it’s the business side of things that’s really been making headlines. The company’s valuation and partnerships tell a story of massive confidence in the future of humanoid robots.
A valuation built on future promise
Figure AI’s financial growth has been wild. After a funding round in early 2024 valued the company at $2.6 billion, it’s now rumored to be in talks for another round that could push its valuation to a staggering $39.5 billion. This number isn’t based on what the company is earning today (which isn’t much), but on the enormous potential of the market down the road.
Investors like Parkway Venture Capital, Microsoft, NVIDIA, and Amazon’s Jeff Bezos are placing a huge bet. They’re looking at forecasts like one from Goldman Sachs, which predicts the humanoid robot market could hit $38 billion by 2035. They don’t just see a hardware company; they see a major player in the next big wave of automation.
Strategic partnerships as proof of concept
To get its robots out of the lab and into the real world, Figure AI has to prove they can handle a real industrial setting. The biggest step so far is its partnership with BMW. The plan is to put Figure’s robots to work at the automaker’s plant in South Carolina, where they’ll start automating some production tasks.
This is a huge deal for proving their concept has legs. Still, not everyone is convinced. Some robotics fans and industry folks think the collaboration is more of a small-scale research project than the full-blown factory takeover the hype suggests. For now, it’s a critical first step, but the company has a long way to go to show its robots are ready for prime time.
Reality check: The challenges and skepticism
For all the excitement, Figure AI is trying to solve some incredibly difficult problems, and there’s a good amount of skepticism out there. To get the full picture, you have to look at the tough questions that experts and investors are asking.
Are the demos just "PR-ware"?
One of the biggest knocks against Figure AI, and the whole humanoid robotics field, is that those slick video demos might not be showing the full story. Robotics experts on forums like Reddit often point out that these demos happen in perfectly controlled, scripted environments. Sure, a robot might make a cup of coffee flawlessly on camera, but that could be the one good take out of thousands of attempts. That’s a far cry from an autonomous robot that can deal with the messy, unpredictable nature of a real kitchen or factory. Figure AI still needs to prove its robots are the real deal, not just clever "PR-ware."
Technical hurdles and intense competition
Building a humanoid robot that’s actually useful and affordable is a massive engineering feat. Things like battery life, how much the robot can carry, how fast it can move, and, most importantly, how to make it safe around people are all huge challenges. The hardware has to be capable, incredibly reliable, and cheap enough to make at scale.
And Figure AI isn’t the only one in this race. The competition is getting fierce, with some of the biggest names in tech throwing their hats in the ring.
| Feature | Figure AI (Figure 02/Helix) | Tesla (Optimus) | Boston Dynamics (Atlas) |
|---|---|---|---|
| Primary Focus | General-purpose labor (industrial first) | General-purpose labor (Tesla factories first) | R&D, dynamic mobility, acrobatics |
| AI Approach | In-house vision-language-action models | In-house AI (leveraging FSD data) | Advanced control systems, now exploring AI |
| Commercial Status | Early commercial pilots (e.g., BMW) | In development, internal factory deployment planned | Not commercially sold, R&D platform |
| Known For | Fluid, AI-driven task execution | Potential for massive scale via Tesla manufacturing | Unmatched agility and dynamic movement |
Is the humanoid form factor a dead end?
While Figure AI is betting the farm on the human form, some AI experts wonder if it’s even the right path. Stanford professor and AI pioneer Fei-Fei Li has argued that for a lot of jobs, specialized robots, like ones with wheels or stationary arms, are way more energy-efficient and cheaper. A humanoid robot might be a jack-of-all-trades, but in industrial automation, being a master of one specific task is often what really matters. The debate is still open on whether a general-purpose body can ever beat a specialized machine.
How much does Figure AI cost?
So, what’s the price tag on one of these futuristic workers? That’s a question Figure AI isn’t answering. If you go to the pricing page on its website, you’ll hit a clever 404 "lost in the future" error page.
This isn’t just a cute joke; it points to a classic enterprise sales strategy. The price is probably custom for every customer, based on how many robots they need, how complex the setup is, and how much support they want. This usually means contracts in the six or seven-figure range, long sales processes, and required demos before you even get a price. For most businesses, not knowing the cost upfront is a huge barrier.
Is Figure AI the future or just hype?
Figure AI is without a doubt a super ambitious, well-funded company pushing the limits of robotics. The progress it has made in just a couple of years is impressive, and it has definitely caught the public’s imagination. But it’s also floating in a bubble of massive valuations, intense hype, and some truly monumental challenges. The idea of a humanoid workforce is exciting, but these machines are realistically still years, maybe even a decade, away from being a common and affordable sight. For now, Figure AI is a fascinating, high-stakes gamble on what the future of work might look like.
Get started with practical AI today
While humanoid robots are still figuring out the physical world, your business has problems you can solve with AI right now. You don’t have to wait for a robot to start automating repetitive work, answering common questions, and helping out your team.
With eesel AI, you can get a powerful AI agent running in your helpdesk in just a few minutes. It connects to your old support tickets, documents, and knowledge bases to learn how your business works. Then, it starts handling frontline support, drafting on-brand replies for your agents, and sorting tickets automatically. It’s a practical and affordable way to use AI to solve real problems today, no robot required.
A screenshot of the eesel AI platform showing how the AI agent connects to past support tickets to build its knowledge base, a practical alternative to Figure AI for business automation.Frequently asked questions
Figure AI aims to address labor shortages by developing general-purpose humanoid robots that can seamlessly integrate into human-designed environments like factories and warehouses. Their goal is to have robots perform dangerous or tedious tasks alongside humans.
Figure AI combines advanced hardware with its proprietary vision-language-action (VLA) neural network, Helix. This AI allows the robots to perceive, understand commands, and perform complex physical tasks, learning by observation rather than strict programming.
The humanoid form is strategic because our world is built for humans. This design allows Figure AI robots to navigate existing environments, use standard tools, and integrate into current workflows without requiring costly infrastructure redesigns.
Major investors include NVIDIA, Microsoft, Jeff Bezos (Amazon), and Parkway Venture Capital. Figure AI has also partnered with BMW for early commercial pilots, integrating their robots into a manufacturing plant.
Significant challenges include proving robot capability beyond controlled demos, overcoming technical hurdles like battery life and safety, and intense competition. Some experts also question if the general-purpose humanoid form is the most efficient solution for industrial tasks.
Figure AI does not publicly list pricing for its robots. Pricing is custom and likely determined through an enterprise sales process, involving demos and potentially long contracts in the six or seven-figure range, depending on customer needs.
While competitors like Tesla and Boston Dynamics also pursue humanoid robotics, Figure AI’s primary focus is on general-purpose labor for industrial applications, driven by its in-house vision-language-action models for fluid task execution.
