Humanoid robots are "only heard of in orders, but never seen in deliveries."

The phenomenon of "self - dealing" orders in the robotics industry is not a simple opposition between "good and bad", but a complex product of a specific stage of industrial development. From the experience of the new energy vehicle industry, related orders have an irreplaceable "phased value" in technology verification and production capacity ramping up. However, when the industry enters the stage of large - scale commercialization, it must return to the essence of "real demand".

Orders worth tens of millions or even hundreds of millions are pouring in, and humanoid robots have received positive feedback for their development.

The order wave is not just a short - term boost on the balance sheet, but a crucial leap into a virtuous development cycle. The funds from orders can be used for further technological iteration, and the corresponding application scenarios also provide an excellent stage for the practical testing and optimization of technology, promoting robots to continuously progress from "being able to do" to "doing well". These orders have positive significance that cannot be ignored, whether for the growth of individual enterprises or the cultivation of the entire industrial ecosystem.

However, we also need to clearly recognize that in this seemingly all - positive order boom, it is not entirely pure. There are some "self - dealing" related orders mixed in it.

For example, a company jointly establishes a subsidiary with other entities, and then the subsidiary purchases its own humanoid robots. Nicely put, it's called a "large - scale order", but less nicely... There are even "orders worthy of scrutiny", which are only framework agreements, and there is no sign of delivery for a long time, and the disclosed information doesn't mention the delivery time at all. (Party A: Just buy it, I don't need it.)

For a while, the market can only hear about orders, but no robots are in sight.

It should be clear that in the early stage of the industry, we cannot completely reject all related orders and regard all involved enterprises as "connected parties". After all, at this stage, humanoid robots need verification during the ramping - up period and urgently need venues to show their capabilities. Related orders are a way with relatively low trial - and - error costs for both parties. Through cooperation with related parties, enterprises can quickly collect operation data and iterate product solutions, avoiding the technical adaptation risks and market education costs when directly entering an unfamiliar market. In essence, it is a "safety - cushion - style" verification path in the early stage of the industry.

However, if related orders account for too high a proportion and even become the absolute mainstay of an enterprise's order structure, its potential risks are worthy of attention. On the one hand, excessive reliance on related transactions may lead to "imbalanced" technology verification. The demand in related scenarios is homogeneous and difficult to cover the real pain points of the diversified market, which may easily cause the "dulling" of the product's marketability. On the other hand, the performance boost from related orders may create a "false prosperity", blurring the real market demand boundary of the industry and triggering the risk of blind expansion of low - end production capacity and the formation of bubbles.

All in all, related orders are attractive, but they shouldn't make us lose our heads.

Phased Demand: Related Orders Prop Up the "Valley of Death"

"In the early stage of the development of disruptive technologies, mainstream customers never choose to use disruptive products." In The Innovator's Dilemma, looking back at the history of human science and technology industries, almost all the rise of disruptive products has to go through a long cultivation period of "technology exploration - scenario verification - large - scale outbreak". From the laboratory prototype of computers to the desktop terminals in thousands of households, and from the policy - dependent stage of new energy vehicles to their market - wide popularization, emerging industries often face multiple dilemmas in the early stage, such as "immature technology, high costs, and unclear demand". They need continuous R & D investment to break through bottlenecks, feedback from real scenarios to complete iteration, and the support of policies and capital to cross the "valley of death". Today's humanoid robot industry is such a strategic and disruptive field that is highly anticipated but still in its infancy. However, in the product market, ordinary families won't choose humanoid robots in the short term, and in industrial scenarios, they can't compete with traditional automated equipment and robotic arms. Humanoid robots can only find sporadic "marginal orders". This leads to a time - line misalignment between the strategic importance of humanoid robots and the maturity of industrial development, thus delaying the pace of industrial development.

At this time, it is precisely the orders led by non - real commercial forces such as industrial investors that are needed to drive the deployment of humanoid robots in positions and enable them to learn in scenarios. History always repeats itself. We can first look back at the early stage of the new energy vehicle industry. "Self - dealing" related orders were a common phenomenon in the industry at that time. At that time, in the early orders of some car manufacturers, many came from internal group purchases (such as car manufacturers supplying vehicles to their own travel platforms) or locally - led related transactions (such as the public transportation system purchasing local new energy vehicle models). Although these orders bore the mark of internal circulation, they played an irreplaceable role in the early stage: they helped enterprises complete the early production capacity ramping - up, spread out R & D and manufacturing costs, and provided real - scenario data support for technological iteration. This is highly consistent with the logic of related orders in the current robotics industry.

The current robotics industry is at a critical stage of "technology verification - early stage of commercialization". The rationality of related orders is essentially to solve many pain points in industrial development.

From the perspective of market demand, the real consumer demand for humanoid robots has not fully erupted. If ordinary enterprises only rely on market - based orders, they are likely to fall into a survival dilemma of "high R & D investment and low revenue return". The bundled cooperation model of "investment + orders" is equivalent to providing "life - saving funds" for enterprises, allowing technology R & D to continue to advance.

In terms of industry characteristics, the iteration cost of robot products is extremely high. The R & D investment for a single unit often exceeds tens of millions or even hundreds of millions of yuan. The current robot products are not standardized products, and directly launching them into a market with unclear demand poses a huge risk. With the help of framework orders and agreement orders from related parties, enterprises can adjust product parameters and deployment through negotiation, greatly reducing the trial - and - error cost.

From the perspective of ecological positioning, most of the enterprises holding large - scale related orders at present are players that have already occupied a position or are competing for the leading position in the industry. Whether they are preparing for listing, meeting local industrial needs, or building industry influence, the chain of "orders - revenue - valuation - industry status" is the inevitable path for enterprises to achieve their long - term goals, and related orders are the stepping stones on this path.

However, we also need to clearly recognize that the core flow of current orders is still concentrated in the B - to - B or even G - to - B sectors. The key to the real explosion of the industry lies in when C - to - B orders can surpass B - to - B orders. Currently, the C - end market has not been opened up, and the most fundamental reason is the technological bottleneck. That is, robots still have a gap in being "usable" and "easy to use", and their precision, battery life, interaction ability, etc. have not met consumers' expectations. Therefore, related orders have a certain stage - specific inevitability in the short term.

Beware of the "Boiling Frog in Warm Water" Effect

The phased significance of related orders reflects the industry's confidence in the robotics industry with a long - term perspective. However, this does not mean that we should ignore the risks behind them. When related orders deviate from the essence of "technology verification and industrial paving" and become tools for enterprises to whitewash their performance and inflate their valuations, the systematic harm to the entire industry will become apparent.

Some enterprises in the new energy vehicle industry focused on "manufacturing cars to get subsidies" and invested funds in expanding production capacity rather than core technology R & D. Eventually, after the subsidy period ended, they faded out of the market due to backward technology and lack of market competitiveness.

Now, a similar sign has emerged in the robotics industry. Some enterprises pay more attention to increasing revenue and inflating valuations through related transactions, but invest insufficiently in the R & D of core components and the construction of self - financing capabilities. There is even a tendency to "emphasize the quantity of orders and neglect technological breakthroughs", which goes against the long - term technology - accumulation nature of the industry.

What is even more worthy of attention is that if related orders become the mainstream in the industry, a competitive environment of "bad money driving out good money" may be formed. Enterprises that focus on market - based orders and high R & D investment may have relatively slow revenue growth, and their valuations may be much lower than those of enterprises relying on related orders. As a result, capital will flow to enterprises that are "good at getting orders" rather than "good at technological innovation", ultimately delaying the pace of technological breakthroughs in the entire industry.

We must be vigilant against the industry falling into the misunderstanding of "paying too much attention to appearance and lacking substance" and understand that the core value and advantages of an enterprise are far more than the positive news of a few related orders.

Of course, the harm of false prosperity is also reflected in multiple aspects: financially, excessive reliance on related orders may lead to "bloated" revenue and even hide the risk of fraud. If the related parties have insufficient payment ability, enterprises will also face cash - flow pressure; technologically, easily obtained related orders will weaken the innovation motivation of enterprises, leading to stagnation in technological iteration. After all, "being fed on a silver platter" won't cultivate self - reliance; in terms of market judgment, false order data may mislead the industry's understanding of demand, causing enterprises to fall into "order inertia" and lose the ability to explore the real market.

The phenomenon of "self - dealing" orders in the robotics industry is not a simple opposition between "good and bad", but a complex product of a specific stage of industrial development. From the experience of the new energy vehicle industry, related orders have an irreplaceable "phased value" in technology verification and production capacity ramping up. However, when the industry enters the stage of large - scale commercialization, it must return to the essence of "real demand".

This article is from the WeChat official account "Embodied Learning Society". The author is Peng Kunfang, and the editor is Lü Xinyi. It is published by 36Kr with permission.