Don't want to be trapped in the automotive cycle, Hesai bets on spatial intelligence.

```

Author | Chai Xuchen

Editor | Zhou Zhiyu

While domestic car manufacturers are still fighting price wars over “equal access to smart driving,” lidar companies have already begun searching for their next ticket.

On April 17, Hesai launched its 6D full-color lidar ultra-sensitive chip and announced a strategic upgrade from “spatial perception” to “spatial intelligence,” aiming to equip machines not only to see their surroundings, but also to understand spatial relationships and act autonomously.

Following this trend, Hesai also unveiled Kosmo, a spatial intelligence hardware for the consumer market, as well as a robotics power module, considered a core foundation for the physical AI era.

This company, previously deeply tied to the automotive industry chain, is now urgently seeking to transform its identity.

In the past few years, Hesai benefited from the rise of new energy vehicles in China. Ideal, XPeng, NIO, Great Wall, Leapmotor, and other automakers densely equipped their vehicles with lidar, rapidly helping Hesai become one of the global shipment leaders, and convincing the capital market it is the “shovel seller” in the smart car era.

But the shovel-selling story also has a ceiling.

The automotive market is ultimately cyclical; car makers drive down prices, platform procurement, and divergent smart driving strategies all squeeze supply chain profit margins. As lidar transitions from a high-end feature to a standard in mid-range models, the industry’s competitive logic shifts from “technical scarcity” to “cost efficiency.”

Against this backdrop, it’s no surprise Hesai is looking beyond cars.

From Selling Lidar to Selling Spatial Understanding

The lidar industry once enjoyed a golden era.

At that time, the market believed autonomous driving required high-precision 3D perception, and lidar was the optimal solution. Many startups flooded the track, and valuations soared. Hesai, Huawei, RoboSense, and Tuitu.ai all rose rapidly.

But after a few years, the reality became more complex.

On one hand, Tesla’s approach proved pure vision could achieve a certain level of advanced assisted driving; on the other hand, Chinese automakers quickly made lidar just another line item on the spec sheet. Products once priced at tens of thousands of yuan are now in the thousand-yuan range.

A head of smart driving at a major OEM once said bluntly: “Today, when purchasing lidar, the first criterion is no longer maximum detection range, but BOM cost.”

This is pressure for all suppliers.

Hesai does have its moat. It boasts strong mass production capabilities, self-developed chip technology, and top-tier customer resources. But even the strongest supplier can hardly escape the reality of automakers' tough price negotiations.

So Hesai began telling a bigger story: spatial intelligence. At this launch, CTO Xiang Shaoqing said, “Spatial perception is only the first step—the truly important thing is letting machines understand, operate, and serve space.”

In other words, it’s going from selling hardware to selling system capabilities. If lidar used to “see the world,” future spatial intelligence products will solve “what to do after understanding the world.”

Kosmo is the first product to land. Xiang Shaoqing said Kosmo can fully capture and record the 3D world, and with continuous iteration, may eventually become a “pocket camera” for a broader user base. Meanwhile, it can support physical AI training, offer spatial data to robotics companies for world model training.

An industry analyst told Wallstreetcn: “No matter how big the automotive lidar market gets, it’s essentially dozens of car company clients. But if consumer electronics takes off, it’s a hundreds-of-millions terminal market. These are two different scales.”

Hesai clearly sees this difference.

Why Now

Hesai’s pivot at this time is driven by the marginal slowdown of automotive market growth dividends.

New energy vehicles are still growing, but supply chain profits are being continually compressed. Automakers, competing for market share, transfer price war pressure layer by layer to parts suppliers. An industry insider told Wallstreetcn that although lidar penetration is rising, average selling prices keep falling, making profits from unit price increasingly difficult.

With the maturation of lidar technology, its application scenarios are expanding.

An industry investor told Wallstreetcn that lidar used to be expensive due to complex mechanics, high chip costs, and low mass production yields. Now, with chipification, solid-state designs, and scaled manufacturing, costs are falling much faster than expected.

A person close to Hesai told Wallstreetcn that the company’s core belief in recent years is: lidar is not the end, spatial data is the end. “If we only make sensors, we’ll ultimately enter price competition; if we can master spatial understanding and interactive access, the imagination space is totally different.”

The Picasso SPAD-SoC Hesai launched integrates capabilities previously scattered across multiple modules. After chipification, products are smaller, lower power, more cost-controllable, and better suited to consumer scenarios.

A supply chain insider said: “Many technologies start serving high-end industries, and eventually sink into mass markets. Lidar is at that stage now.”

Plus, AI provides a new entry point. Without large models, spatial perception is mostly hardware upgrades; but with generative AI and embodied intelligence, machines need real-time 3D world understanding, redefining lidar’s value. Robots need obstacle avoidance, grasping, navigation; AR glasses need spatial modeling; home assistants need room structure understanding; autonomous devices need to recognize human-object relationships.

These are not tasks traditional cameras can easily accomplish alone. Hesai’s management stated onsite: “Every intelligent agent in the future will need reliable spatial input.”

In other words, Hesai wants to upgrade lidar from a car part to foundational hardware in the AI era.

Consumer Electronics is Not an Easy Battle

The problem is, consumer electronics has never been a market that suppliers can easily win.

Auto industry cycles are long and certifications strict, but once you’re in the supply chain, orders are stable, and product life cycles are long. Consumer electronics is very different: rapid iteration, aggressive pricing, high brand concentration, and any misstep can lead to substitution.

More importantly, consumers don’t buy “because it’s lidar”—they buy for the experience.

Meaning: If Hesai wants to enter the C-side, it can’t just shrink vehicle-grade lidar and sell it—it must prove it delivers visible, buyable, new experiences.

For example, home robots precisely navigate in darkness; AR devices achieve centimeter-level spatial positioning; smart security truly recognizes abnormal behavior instead of false alarms; companion robots understand people’s positions and actions.

If this isn’t achieved, lidar is likely just an expensive add-on.

A consumer electronics investor told Wallstreetcn: “Many supply chain companies mistakenly believe strong technology means they can succeed in consumer markets. But the consumer market is about product definition, channel capability, and brand mindshare—not the same thing.”

This is the real challenge Kosmo faces.

Hesai’s past strengths are in B2B delivery—stability, reliability, mass production, automotive certification. The consumer market demands aesthetics, interaction, ecosystem, content, and operations.

Transitioning from B2B to B2C is not additive—it requires rebuilding organizational capabilities.

Therefore, rather than directly targeting the mass consumer electronics market, robotics may be a more pragmatic landing point for Hesai. Industry insiders see robotics as combining B2B procurement logic with B2C growth potential.

It needs reliable hardware and values performance specs, but it’s still early-stage, with supply chain structures not yet solidified. Whether it’s warehouse robots, delivery robots, home robots, or companion robots, all need spatial perception.

Hesai’s display of a robotics power module also shows it doesn’t want to be just the “eyes,” but also the “limbs.”

Because if you only make sensors, you’re still just a parts supplier; if you enter actuation systems, you’re closer to a robotics platform company. In a future robot, perception, motion, and decision systems will collaborate, making the machine’s value far greater than individual sensors.

A robot entrepreneur told Wallstreetcn: “Whoever can provide standardized modules for both perception and motion has a chance to be Tier1 in the robotics era.”

Hesai clearly wants to compete for this position and reevaluate itself. Plus, the robotics industry’s competition hasn’t become as brutal as automotive supply chains. No stable barriers among top players yet, and technology routes are changing fast, leaving a window of opportunity for latecomers.

Going forward, whether Hesai can break out depends on two things: First, can Kosmo find a killer scenario, not just remain a technology demo? Second, can the company upgrade from a component mindset to a platform and product mindset?

The new script is just beginning. Regardless of success or failure, Hesai has shown its attitude: it doesn’t just belong to cars. If it succeeds, Hesai’s boundaries will be thoroughly opened.

Risk Disclosure and DisclaimerThe market has risks; investments require caution. This article does not constitute personal investment advice and does not consider individual users’ special investment goals, financial situations, or needs. Users should determine whether any opinions, views, or conclusions in this article are suitable for their specific situation. Investing based on this is at your own risk. ```