The next dominant force in robotics might not be the company making headlines with flashy AI announcements. While Nvidia captures investor attention with its Rubin platform and Alpamayo autonomous AI system, a Chinese lidar manufacturer is quietly controlling the infrastructure that every robot actually needs to see. Hesai Group holds 60-70% market share in robotaxi lidar and dominates the robotics vision segment with 61% of that market specifically. The company shipped 40,000-50,000 units of lidar per quarter in 2025 alone, and has committed to doubling production capacity to over 4 million units annually by 2026—a move that mirrors Nvidia’s own evolution from a graphics card maker to the de facto infrastructure vendor for AI.
The parallel is striking. Just as Nvidia became indispensable to everyone building large language models not by making the models themselves but by providing the chips they run on, Hesai is becoming the foundational supplier for the robotics wave. When Nvidia’s Jensen Huang declared at GTC 2026 that “every industrial company will become a robotics company,” he was describing a future that depends entirely on sensor availability. Hesai is the company actually enabling that future.
Table of Contents
- Why Sensor Dominance Matters More Than AI Software
- Production Ramp as a Competitive Moat
- The Nvidia Misdirection in Robotics Strategy
- The Emerging Chip Competition for Physical AI
- Why Wall Street Hasn’t Noticed Hesai
- The Margin Arbitrage Between Automotive and Robotics Lidar
- What Comes Next for Robotics Infrastructure
- Conclusion
Why Sensor Dominance Matters More Than AI Software
The robotics industry is experiencing explosive growth that most investors haven’t fully internalized. Global robotics funding reached $27.6 billion in 2025, a doubling from $13.7 billion in 2024. Companies like Skild AI raised $1.4 billion in early 2026, and Mind Robotics raised $500 million in a Series A round in March 2026. This represents genuine industrial momentum, not speculative hype. Every one of these startups, along with established robotics manufacturers like ABB, Fanuc, and Boston Dynamics, needs lidar sensors to give their robots spatial awareness. That’s where Hesai’s advantage becomes irreplaceable. Lidar isn’t glamorous.
It doesn’t make earnings call headlines or attract venture capital headlines the way AI software does. But a robot without working lidar is a dangerous brick. Manufacturing robots need lidar to detect obstacles and workers. Autonomous mobile robots in warehouses need it to navigate. Delivery robots need it to avoid collisions. The sensor is the foundation that everything else—the AI models, the control software, the business logic—sits on top of. When one company controls 60-70% of the market for a critical component, it creates enormous pricing power and margin opportunity. Hesai’s 2026 robotics lidar volume is expected to double compared to 2025, and the margins on these shipments are significantly higher than what the company earned from automotive lidar sales.
Production Ramp as a Competitive Moat
What makes Hesai’s position even stronger is that capacity constraints have historically limited the entire robotics industry. If you’re building autonomous robots and you can only source a limited number of sensors, you’re limited in how many units you can manufacture. this has been a chronic bottleneck, but Hesai’s decision to double production capacity removes that constraint for its customers. Moving from 2 million to over 4 million annual units is not a modest upgrade; it’s the difference between scarcity and abundance. The timing is deliberately aggressive. Hesai announced this expansion in January 2026, positioning itself to capture market share precisely when venture funding is flowing into robotics and established manufacturers are ramping up their automation initiatives.
The quarterly volume data—40,000-50,000 units per quarter in 2025—shows the company is already operating at meaningful scale. By doubling capacity, Hesai is ensuring that supply won’t be the limiting factor for its customers, which cements customer loyalty. Companies don’t switch sensor suppliers lightly once manufacturing pipelines are established. There is a real risk here, however: if demand doesn’t materialize as expected, Hesai could end up with excess capacity and margin compression. The robotics funding boom is real, but execution risk remains. Early-stage companies may fail, larger manufacturers may move slower than anticipated, or competing lidar technologies could improve faster than expected.
The Nvidia Misdirection in Robotics Strategy
Nvidia is playing a longer game in robotics through its Rubin platform, a suite of six specialized chips designed for autonomous machines. The company plans to make these available to partners starting in Q2 2026. Additionally, Nvidia launched Alpamayo, marketed as “the world’s first thinking, reasoning autonomous vehicle AI,” which suggests the company is building a full stack from chips to software. This makes Nvidia look like the company to watch in robotics, and the narrative is compelling to investors. Nvidia’s established position, enormous R&D budget, and relationships with major manufacturers like Toyota and Hyundai suggest the company will play a significant role.
But here’s the critical distinction: Nvidia is competing in the compute layer—the silicon that processes data and runs AI models. Hesai competes in the sensor layer—the hardware that actually captures the data in the first place. These are different businesses with different dynamics. A robot needs both Nvidia compute and Hesai sensors, but you can theoretically substitute Nvidia chips with hardware from AMD, Qualcomm, or SambaNova (which just launched the SN50 chip in February 2026, claiming 5x faster speed for agentic AI and 3x lower total cost of ownership). You cannot easily substitute a 60-70% market share lidar sensor supplier. That’s a structural advantage that’s harder to disrupt through superior technology or aggressive pricing.
The Emerging Chip Competition for Physical AI
While Hesai maintains clear dominance in lidar, the semiconductor layer is fragmenting in ways that could reshape robotics economics. SambaNova’s new SN50 chip represents a genuine technical challenge to Nvidia’s assumed superiority in robotics AI inference. If companies can run physical AI workloads faster and cheaper on alternative chips, it changes the economics of building robots. Positron raised over $230 million in a Series B round in early 2026 with backing from Arm Holdings, QIA, and Jump Trading, suggesting institutional investors see real differentiation potential outside of Nvidia’s stack.
This matters for cost-sensitive applications. A delivery robot company operating thousands of units is extremely sensitive to the per-unit cost of compute. If Positron or SambaNova can deliver equivalent performance at lower cost, switching is economically rational. Hesai, meanwhile, maintains a different kind of moat: they’re already in production at massive scale, the lidar sensor is expensive enough that customers won’t switch for marginal improvements, and the manufacturing relationship creates switching costs. The robotics companies that succeed will likely need to be agnostic about compute while being strategically dependent on their lidar supplier.
Why Wall Street Hasn’t Noticed Hesai
If Hesai Group controls 60-70% of a critical component in a robotics market experiencing 101% funding growth, why isn’t it a household name among investors? The answer reveals something important about how markets work and what gets priced in. Hesai is a business-to-business supplier. It doesn’t sell directly to consumers, doesn’t have a branded product that anyone can touch or understand immediately, and doesn’t make grand proclamations about transforming industries. The company’s investors are mostly Chinese VCs and strategic investors from automotive OEMs that already relied on Hesai for autonomous vehicle sensors. The lack of Wall Street attention also reflects a geographic bias.
Western investors are accustomed to betting on American or European robotics companies. Hesai’s primary market has been China, both for robotaxi lidar and increasingly for domestic robotics. As the company expands globally and production capacity increases, Western investors will eventually notice. But by then, much of the optionality will be gone—Hesai will already control the supply chain. This is the “hiding in plain sight” dynamic referenced in the title: the most important company in robotics infrastructure is already dominant, but it’s not on anyone’s list of robotics stocks to watch.
The Margin Arbitrage Between Automotive and Robotics Lidar
A concrete advantage Hesai holds is that robotics lidar carries higher margins than automotive lidar, even though it’s fundamentally the same technology. This matters enormously for shareholder returns. In the automotive market, intense competition between suppliers and the cost-sensitivity of car manufacturers meant lidar suppliers had to accept razor-thin margins.
In the robotics market, a $100 reduction in lidar cost might not change whether a company builds a robot, but a 40-50% lidar supplier margin looks extremely attractive compared to the 10-15% margins typical in automotive. As Hesai’s business mix shifts from automotive to robotics—with robotics lidar volume expected to double in 2026—the company’s overall profitability should improve even if total unit volumes grow only modestly. This arithmetic is already understood by institutional investors in Asia, which is likely why companies like Bank of America have named Hesai a top pick in robotics. The lesson for Western investors is that the most profitable robotics companies may not be the ones making the robots or the AI software, but the ones supplying the irreplaceable components that sit between sensing and decision-making.
What Comes Next for Robotics Infrastructure
The next phase of the robotics boom will reveal the true winners and losers in the infrastructure layer. If the robotics funding surge translates into actual robot deployments over the next 2-3 years, sensor demand will become the binding constraint on industry growth. Companies that secured long-term supply agreements with Hesai, or that diversified their sensor suppliers early, will have a significant advantage. Companies that waited, assuming they could source sensors off-the-shelf at commodity prices, will face delays and cost pressures.
The competitive threat to Hesai is real but not imminent. SambaNova, Positron, and others may disrupt the compute layer, but that doesn’t threaten the lidar segment. A potential future competitor in lidar could emerge from established companies like Sick AG or Velodyne, or from new entrants with alternative sensing technologies like solid-state lidar. But none of these threats are current, and Hesai’s production ramp—moving from 2 million to 4 million annual units—suggests the company is planning to consolidate its dominance before meaningful competition arrives. By 2027 or 2028, when robotics deployments accelerate, Hesai’s grip on the industry may be difficult to challenge purely on cost or performance grounds.
Conclusion
The narrative around robotics has been dominated by big names: Nvidia and its Rubin platform, Tesla and its humanoid robots, Boston Dynamics and its impressive demos. These companies matter, but they’re not the hidden Nvidia in robotics. That title belongs to Hesai Group, a lidar sensor manufacturer that controls 60-70% of a market segment experiencing 101% annual funding growth. The company is doubling production capacity, experiencing margin expansion as its business mix shifts from automotive to robotics, and operating at a scale that makes displacement difficult.
The deeper insight is that infrastructure companies—the suppliers of critical, irreplaceable components—tend to be more valuable and more durable than the flashy companies building end products. Nvidia understood this in AI. Hesai is now operating from the same playbook in robotics. For investors and robotics companies trying to understand where the next decade of value creation will come from, tracking Hesai’s capacity ramp and market share dynamics will be more predictive than tracking which startup just announced a new robot prototype.
