Several investors are now eyeing robotics companies with the same conviction they once had in Nvidia, seeing similar forces at play: an emerging technology wave, massive market expansion, and a company positioned to supply the enabling infrastructure. The reasoning is straightforward—just as Nvidia captured value by providing the GPUs that power every AI model and data center, early robotics plays could capture disproportionate value by providing the AI chips, software platforms, and foundational technologies that hundreds of manufacturers will depend on. Figure AI’s $39 billion valuation and Nvidia’s own $586 million in robotics revenue during fiscal 2026 suggest the market is taking this thesis seriously.
The comparison hinges on a critical similarity: infrastructure plays tend to outperform end products in fast-growing industries. During the AI boom, Nvidia’s valuation exploded not because the company made the best chatbot or language model, but because every organization building those applications needed Nvidia hardware. If robotics evolves along a similar trajectory—with thousands of manufacturers deploying humanoid and automated systems—then the companies controlling the underlying compute, software stacks, and AI models could see Nvidia-scale returns. The global robotics market is projected to grow from $73.64 billion today to $218.56 billion by 2031, a 19.86% compound annual growth rate that creates the kind of expanding addressable market that justifies billion-dollar valuations.
Table of Contents
- Why Robotics Could Mirror AI’s Infrastructure Opportunity
- The Companies Investors Are Watching for “Next Nvidia” Status
- Technology Moats and Why Software Beats Hardware
- Valuation Comparisons and the “Next Nvidia” Premium
- Manufacturing Scale and the Execution Risk Overlooked
- Real-World Deployment and Early Wins
- The Path to Dominance and Future Outlook
- Conclusion
Why Robotics Could Mirror AI’s Infrastructure Opportunity
The appeal of robotics for investors stems from the sheer scale of the addressable market and the structural advantages of being an infrastructure provider rather than a hardware manufacturer. When Nvidia became dominant in AI, it wasn’t by building applications—it was by becoming the bottleneck that no one else could replicate at scale. Similarly, robotics companies that control the AI models, compute modules, and software platforms stand to capture more value than robot manufacturers themselves, much like how GPU makers prospered more than individual AI startups. Boston Dynamics announced in January 2026 that its Atlas humanoid robot is entering commercial production, with parent company Hyundai planning to manufacture 30,000 robots annually by 2028. When a single company is deploying tens of thousands of units, and each unit requires compute, vision processors, AI models, and software updates throughout its lifetime, the infrastructure supplier becomes indispensable.
Nvidia’s Jetson T4000 module, powered by Blackwell architecture, delivers 1,200 teraflops of AI compute in just 40-70 watts—a specification tailored specifically for the robotics use case. this isn’t incidental; it’s evidence that major chip manufacturers are building products with robotics as a primary market. However, the robotics opportunity differs from AI in one crucial way: manufacturing and deployment complexity. Nvidia sold software and chips to software engineers who could immediately deploy them. Robotics requires physical manufacturing, supply chains, field service, and warranty support. This means robotics infrastructure companies still face capital intensity and execution risk that pure-play chip companies avoided, which could dampen returns compared to the AI analogy.
The Companies Investors Are Watching for “Next Nvidia” Status
Figure AI stands out as the most talked-about candidate, having raised funding at a $39 billion valuation in its Series C round with participation from Nvidia itself—a signal that even Nvidia sees Figure as a potential long-term partner or acquisition target. Figure’s focus on building advanced humanoid robots and the software to control them mirrors how Nvidia moved from pure GPUs to software stacks like CUDA that locked in developer and customer loyalty. Boston Dynamics, while owned by Hyundai, remains a critical player in the robotics infrastructure conversation. Its Atlas robot is designed to be a general-purpose platform, and the company is licensing its technology and control software to other manufacturers.
The fact that Hyundai plans 30,000 annual units by 2028 suggests Boston Dynamics may follow the google Maps model—achieve critical mass in your own operations, then monetize the underlying platform for others. Tesla, often overlooked in robotics conversations despite decades of manufacturing automation expertise, is building proprietary robotic systems for its factories and is exploring broader commercialization of humanoid robots through its Optimus program. The real “next Nvidia” candidate, though, may not be a single robot manufacturer but rather a company controlling the foundational technology layer. Nvidia itself is already positioning its robotics segment aggressively, having generated $586 million in revenue from its automotive and robotics division in Q2 of fiscal 2026. The company’s Isaac GR00T N1.6 model, unveiled at CES 2026, is designed to allow humanoid robots to understand and execute complex instructions—a key capability that every future robotics manufacturer will need to license or purchase.
Technology Moats and Why Software Beats Hardware
Infrastructure companies in robotics will build value through three mechanisms: proprietary AI models, specialized compute hardware, and software ecosystems that competitors struggle to replicate. Nvidia demonstrated this with CUDA, a software platform that made its GPUs the default choice for machine learning—even when competitors produced comparable hardware, CUDA’s ecosystem lock-in maintained Nvidia’s dominance. In robotics, we’re already seeing analogous layers forming. Nvidia’s Isaac platform provides roboticists with simulation tools, AI models for vision and reasoning, and ready-made software modules for navigation and object manipulation. Companies using Nvidia’s stack include Boston Dynamics, Caterpillar, Franka Robotics, LG Electronics, and NEURA Robotics.
When that many manufacturers depend on your software platform, you can raise prices, update licensing terms, and ensure that every new product category (legged robots, mobile manipulators, collaborative arms) becomes dependent on your ecosystem. tesla and Boston Dynamics, by contrast, are betting that proprietary, vertically integrated systems will outperform—a riskier proposition because they must excel at both hardware and software simultaneously. The limitation here is that software moats in robotics are less durable than in software-only businesses. A robot manufacturer unhappy with Nvidia’s licensing can invest in custom silicon and competing AI models—it’s expensive and time-consuming, but it’s possible. Nvidia’s moat relies on continuous innovation and ecosystem depth, not on any single insurmountable technical barrier. If a competitor achieves parity in AI performance while offering better price-to-performance, switching costs may be lower than expected.
Valuation Comparisons and the “Next Nvidia” Premium
Investors comparing robotics companies to Nvidia often cite historical multiples as justification for current valuations. Nvidia trades today at roughly 36 times forward earnings, a premium that reflects both past AI growth and expected future dominance. Figure AI’s $39 billion valuation is being evaluated on similar reasoning—if it captures even a fraction of the robotics infrastructure opportunity, current pricing could look cheap a decade from now. The issue is that “similar opportunities” rarely produce “similar outcomes.” Robotics faces headwinds that AI did not: manufacturing complexity, regulatory approval cycles, and the fact that Nvidia is already competing in robotics directly. When Nvidia entered AI, it had few credible competitors in specialized compute hardware.
Today’s robotics startups face Nvidia as both a potential customer and a competitor, a dynamic that limits their leverage and potential margins. Symbotic, for example, has built an AI-powered supply chain robotics platform and secured a multi-year partnership with Walmart—impressive traction, but the company remains private and much smaller than Figure AI, illustrating how difficult it is to scale in an industry with massive execution requirements. A more conservative valuation approach would compare emerging robotics companies to Caterpillar (a diversified manufacturer with strong margins) rather than Nvidia (a pure-play technology company). Caterpillar trades at roughly 12 times forward earnings, a fraction of Nvidia’s multiple. The truth likely lies in the middle: a successful robotics infrastructure company could command a premium to industrial manufacturers but may never reach Nvidia’s current valuation multiples.
Manufacturing Scale and the Execution Risk Overlooked
One reason investors are enthusiastic about robotics companies is the market growth projection—from $73.64 billion in 2025 to $218.56 billion by 2031 represents a 19.86% compound annual growth rate, which is substantial but not unprecedented for manufacturing-adjacent industries. The real risk is execution: turning that growth forecast into actual unit sales and customer satisfaction. Nvidia, in its early decades, faced minimal execution risk because software and chips ship at near-zero manufacturing cost. Robotics companies must actually build and deliver physical products that work reliably in customer environments. Boston Dynamics’ Atlas entering production is a major milestone, but the company must now sustain manufacturing quality, manage supply chains for components, train customer workforces, and handle warranty claims at scale.
A 30,000-unit annual production run is ambitious for a robot that costs hundreds of thousands of dollars and requires on-site support. Any major defect or safety issue could halt growth and crater valuations. Historical precedent is instructive here: Amazon Robotics (formerly Kiva Systems) is extraordinarily successful in warehouses because the robots operate in controlled environments with minimal human interaction. Humanoid robots operating in factories alongside workers face far greater complexity and liability. This execution risk makes early-stage robotics companies fundamentally riskier investments than Nvidia was during its AI expansion, suggesting that “next Nvidia” returns may remain speculative until manufacturers actually deploy tens of thousands of units successfully.
Real-World Deployment and Early Wins
The most concrete evidence supporting the “next Nvidia” thesis comes from actual deployments. Boston Dynamics has announced that Atlas units will deploy to Georgia factories for parts sequencing, marking the transition from research to production use. This is significant because it validates the fundamental premise: humanoid robots can perform useful work in real manufacturing environments, and manufacturers are willing to invest in them at scale.
Hyundai’s plan to manufacture 30,000 robots annually by 2028 is the largest production commitment we’ve seen to date. That scale is large enough to move the needle on supply chain demand, create network effects, and justify vendor ecosystem development. When a single customer plans that volume, software platforms, component suppliers, and logistics networks all suddenly become viable businesses. This is the moment where a “next Nvidia” really emerges—not from theoretical opportunity, but from an existing customer’s commitment to scale.
The Path to Dominance and Future Outlook
If any robotics company achieves “next Nvidia” status, it will likely follow Nvidia’s playbook: lead with technology that’s demonstrably superior, build a developer and manufacturer ecosystem around it, establish industry standards, and then continuously innovate to stay ahead. Figure AI and Nvidia itself are both pursuing this strategy, while Tesla is betting on vertical integration.
Looking ahead, the robotics market will likely consolidate around 2-3 dominant infrastructure providers, much like how GPU compute consolidated around Nvidia. The companies that win will be those that solve the cold-start problem—getting the first tens of thousands of units deployed reliably, then using that traction to attract more manufacturers and developers. With Boston Dynamics entering production, Hyundai backing humanoid robots at scale, and the global market projected to triple by 2031, the moment of rapid consolidation may be closer than investors realize.
Conclusion
The case for a “next Nvidia” in robotics rests on two solid foundations: the sheer market opportunity (growing at nearly 20% annually toward a $218 billion market by 2031) and the structural advantage of controlling foundational technology layers rather than end products. Figure AI, Boston Dynamics, and Nvidia itself are all positioned to capture this value, though execution risk and manufacturing complexity make outcomes less certain than Nvidia’s AI dominance turned out to be.
For investors, the opportunity is real but requires patience and selectivity. Companies with actual manufacturing partnerships, demonstrated AI capabilities, and capital efficiency are more likely to deliver “next Nvidia” returns than those relying purely on hype or speculative technology. The next five years will reveal which robotics infrastructure companies have built genuine moats versus which ones face commoditization as manufacturing scales.
