Lighting up the Physical Blind Spot of AI: The "Three-stage Rocket" of Passive Internet of Things

The well - known enterprise in the field of passive Internet of Things, Wiliot, posed a thought - provoking question in a recent article: Why can't and shouldn't your AI agent make supply - chain decisions independently?

The article envisioned a scenario where a powerful supply - chain AI agent, based on a vast dataset and precise algorithm models, confidently predicted an imminent surge in market demand and automatically placed a purchase order for 50,000 products.

On the surface, this seemed to be an efficient and rational decision. However, the AI overlooked several crucial real - world factors. It was unaware that this “demand peak” actually stemmed from a short - term market gap caused by a competitor's product recall. It also failed to notice that the production partner's factory was experiencing severe quality - control issues. More importantly, it had no knowledge that the R & D department of its own company was about to launch a brand - new flagship product in the next quarter that would reshape the market landscape.

This case precisely reveals a fundamental contradiction in the current development of artificial intelligence: the gap between powerful computing capabilities and limited perception of the real world.

AI is almost omnipotent in the digital world. But when it attempts to act in the physical world full of dynamic variables and complex feedback, it often makes errors due to the lack of real - time, fine - grained perception. It is like a general who is good at strategizing on a “digital twin” sand table but may misjudge the situation on the real battlefield due to incomplete intelligence.

The new - generation ubiquitous communication and perception technology represented by Ambient Internet of Things (Ambient IoT) is providing AI with the crucial “senses” — enabling algorithms not only to calculate but also to “perceive” the subtle changes in the real world.

This not only means a further integration of communication and perception but also implies that future intelligent systems will no longer be just “thinkers” in the digital space but will become “actors” in the real world. This article will further explore how the Ambient IoT illuminates the physical blind spots of AI and propels us into a stage of deep integration of perception and intelligence.

What are the “nerve endings” for AI to perceive the world?

The cognitive boundary of artificial intelligence ultimately depends on how real and far it can “see”. Even though current large models have shown amazing capabilities in language understanding and reasoning, they are still confined to a fact: AI mainly exists in the digital world, and the dimensions of the information it obtains are still too flat.

The real world is far more complex than data tables. A strategic shift in the board of directors, a secret product iteration in the R & D department, and a temporary adjustment of payment terms by the finance team to optimize cash flow — these key signals that determine the business direction are often beyond the “vision” of AI. Without the perception of these high - level intentions and real - world dynamics, no matter how precise the “optimization” made by AI is, it may go further and further in the wrong direction.

Although traditional IoT technology tries to “equip” AI with “eyes”, its perception is still discrete and fragmented. A temperature reading or a location update is like groping for a corner of the real world in the dark, lacking continuity and context. Decisions made based on these fragmentary signals are likely to be “precisely solving a wrong problem”.

To break through this cognitive blind spot, the industry has proposed the concept of “Physical AI”, advocating that intelligence should not only exist in algorithms and computing power but also be rooted in the real operation of the physical world.

To achieve this, AI needs a new layer of “perception interface” to build a bridge between the digital and physical worlds. One of the basic modules of this bridge is constructed by Ambient Internet of Things (Ambient IoT).

With its characteristics of “ubiquity, continuity, low power consumption, and low cost”, the Ambient IoT provides an unprecedented physical perception foundation for AI, aiming to achieve “99% visibility” of physical assets.

In other words, the objects of perception are no longer just high - value devices but can extend to every pallet, every packing box, and even every commodity in the supply chain... This almost complete panoramic view of the supply chain makes the previously ignored “dark matter” clearly visible.

Of course, “visibility” is not only reflected in the coverage but also in the richness of dimensions. Different from traditional RFID, which can only identify identities, Ambient IoT tags can continuously collect and transmit multi - dimensional data such as temperature, humidity, location, and acceleration, weaving discrete “points” into continuous “lines” and dynamic “surfaces”. AI can thus stop relying on static “data snapshots” and gain insights into a “real - time documentary” of the real world.

What supports all this is the collaborative innovation of the entire industrial chain. From the organic photovoltaic OPV energy harvesting technology jointly developed by Linxens and Dracula in France, to the manufacturing of flexible inlays, and then to the ultra - low - power chips provided by companies such as STMicroelectronics, the entire ecosystem is jointly building a self - sensing and self - powered physical information network.

After AI obtains this continuous, fine - grained physical perception, its role in the business system also changes accordingly. The responsibility of AI is to capture signals in the noise of the physical world, warn of risks before problems occur, and provide deductions and insights for different scenarios; while the responsibility of humans is to understand complex organizational goals, balance multiple interests, and inject experience and intuition. The basis for their collaboration is the “shared physical truth” created by the Ambient IoT.

The three - stage rocket: The value model of Ambient IoT

Recently, a piece of news has caused a sensation in the global technology and retail circles: The global retail giant Walmart announced that it will cooperate with the Ambient IoT pioneer Wiliot to deploy up to 90 million “stamp - sized” smart tags in its supply - chain system.

Meanwhile, Walmart is collaborating with OpenAI, enabling consumers to directly browse and purchase Walmart's products on ChatGPT. The combination of the two can generate considerable commercial value, and the scenario where a home robot butler directly orders goods is within reach.

This is not just a symbolic technology pilot. It is a vote of confidence cast by an industry giant with real money. The Ambient IoT has officially stepped out of the laboratory and entered a new era of large - scale commercialization.

More importantly, Walmart's move is not just to track the circulation of a few pallets. Its real intention is to connect its powerful AI decision - making system to an unprecedented, real - time, and precise physical - world data source.

Behind the implementation of this technology lies the three - fold value evolution of the Ambient IoT: from the “TCO engine” for cost reduction, to the “ESG booster” for supporting sustainable development, and then to the “physical intelligence payload” for symbiosis with artificial intelligence. This is a “three - stage rocket” driving business intelligence, with each stage advancing and the power gradually increasing.

First - stage rocket: TCO engine — Start commercialization from the “economic account”

The Ambient IoT can truly enter the business system because it makes economic sense. For enterprise managers, nothing is more convincing than reducing the total cost of ownership (TCO). According to ABI Research, the shipment volume of Ambient IoT devices will exceed 1 billion in 2030, and the core driving force is the cost revolution brought about by replacing traditional battery - powered solutions.

Enterprises with tens of thousands of tracking devices no longer need to purchase, replace, or recycle batteries, which not only saves hard costs but also a large amount of operation and maintenance manpower. RCR Wireless analysis points out that the “maintenance - free” feature is the most direct and powerful competitive advantage of the Ambient IoT compared with traditional active solutions.

At the same time, the efficiency improvement it brings is not limited to cost - cutting. Take the Royal Mail in the UK as an example. They deployed Ambient IoT tags nationwide to track the circulation paths of letter cages in real - time. This not only avoids asset losses of millions of pounds every year but also optimizes the allocation of dispatch vehicles through real - time data, improving turnover efficiency. The value proposition at this stage is simple and direct: Save, improve efficiency, and be quantifiable.

Second - stage rocket: ESG booster — From “optional” to “essential”

When commercialization passes the early stage, the wave of sustainable development ESG becomes the second driving force for technology popularization. Behind the rapid growth of the Internet of Things lies an often - ignored fact: Approximately 78 million batteries are discarded every day worldwide after being used to power IoT devices, which is a slowly accumulating environmental crisis.

The Ambient IoT uses energy self - harvesting technology to draw energy from the environment, eliminating the need for disposable batteries and fundamentally solving this hidden problem. More importantly, the global regulatory and compliance trend is turning this technology from a “business choice” into a “regulatory necessity”.

For example, the Digital Product Passport (DPP) to be implemented by the EU is an important part of its “Ecodesign for Sustainable Products Regulation”. In the future, products entering the European market must record and track information throughout their life cycle, such as material sources, carbon footprints, and maintenance records. With its low cost and easy deployment, the Ambient IoT is an ideal carrier for implementing this system.

Meanwhile, in an era where ESG has become a key indicator for enterprise valuation, the adoption of Ambient IoT technology is also a clear brand statement. As emphasized in the joint statement of Linxens and Dracula: Sustainability is not just a slogan but an extension of competitiveness. The value proposition at this stage has gone beyond financial logic and shifted to the brand and responsibility logic of enterprises.

Third - stage rocket: The essence of symbiosis with AI — Let the physical world “speak”

The ultimate significance of the Ambient IoT lies not in cost or compliance but in its deep symbiosis with artificial intelligence. This is the real essence of the “three - stage rocket”: Enable AI to have a nervous system for perceiving the physical world.

Currently, AI is suffering from “data hunger”. After the text, pictures, and video data on the Internet have been repeatedly collected and trained, AI needs new high - quality data sources to break through the bottleneck.

The Ambient IoT precisely opens this door, turning trillions of previously silent objects (cartons, clothing, medicines, food) into real - time data nodes and providing AI with the largest “real - world data stream” on the planet.

When these trillions of objects start to “speak”, humans are unable to understand such a vast amount of information. Only AI can act as the “translator”. The Ambient IoT is responsible for perception and collection, and AI is responsible for understanding and reasoning. The two form a closed - loop system — Perception drives cognition, and cognition feeds back to perception.

Walmart's project is the most realistic example of this logic. Deploying 90 million smart tags is not for technological showmanship but to enable the AI system to obtain a continuous stream of real supply - chain data, thus achieving more accurate demand forecasting, more dynamic inventory optimization, and more resilient logistics scheduling.

This marks that the concept of “symbiosis between AI and the Ambient IoT” has moved from a futuristic vision to a real - world reality full of infinite value.

The rainforest ecosystem of Ambient IoT and Eastern wisdom

The future blueprint of the Ambient IoT cannot be realized by the single - point breakthrough of one enterprise or one technology. It is destined to be a cross - industry and cross - field systematic project that requires a complex and healthy industrial ecosystem to support.

At the top of the ecosystem, in February this year, we saw the “Ambient Internet of Things Alliance (AIoTA)” jointly initiated by industry giants such as Qualcomm, Intel, and PepsiCo. Its role is more like the “United Nations” in the IoT world. It does not directly manufacture products but formulates rules of the game. Its core mission is to promote the integration of multiple communication standards such as Bluetooth, NFC, and UWB, ensuring that future Internet - of - Things devices can communicate smoothly in a “common language”.

Facing the layout of European and American enterprises, if China wants to achieve a breakthrough, it neither can “start from scratch” in isolation nor passively follow. Only by adopting a combined strategy of “integration, digestion, and transcendence” can it make a leap from following to leading.

Strategy 1: Integration and voice

It is recommended that Chinese technology giants actively join international mainstream standard - setting organizations such as AIoTA and actively enter the core circle of global standard - setting. Its strategic value lies in two aspects: First, to keep abreast of international technology trends in a timely manner and avoid strategic misjudgments caused by information gaps; second, to let the “Chinese voice” be heard in the rule - making process and incorporate the characteristics and needs of China's industry into the global standard framework, preventing future constraints by standard barriers.

Strategy 2: Digestion and internal circulation

While actively integrating into the international system, China also needs to build and strengthen an industrial closed - loop centered on the domestic market. For example, China Mobile has joined hands with industrial partners to establish the “Ambient Internet of Things Technology Innovation and Industrial Ecosystem Alliance”, aiming to promote in - depth collaboration among “industry, academia, research, application, and investment”, accelerate the large - scale application and industrial upgrading of Ambient IoT technology, and create an open ecosystem and cooperation platform for the Ambient IoT.

Strategy 3: Transcendence and demonstration

China's greatest advantage lies in its super - large - scale market and strong manufacturing and R & D capabilities. We can give full play to this advantage and achieve a technological leapfrog through “application - driven” development. For example, we can learn from the Walmart model, select strategic scenarios, and use concentrated and huge market demand to drive the rapid maturity of the domestic industrial chain, promote cost reduction and technological iteration, define industry standards from the application side, and then achieve a leading position globally.

Conclusion

Looking back at the whole article, the “three - stage rocket” model outlines the evolution path of the Ambient IoT:

• It ignites the spark of commercialization with its TCO advantage.
• It gains policy and capital support for large - scale deployment through the ESG wave.
• Finally, it illuminates the physical blind spots of AI and releases its ultimate disruptive value.

This means that the Ambient IoT is not an extension of traditional technology but a key bridge for the integration of the digital and real worlds. Its value lies in constructing the “shared physical truth”.

This article is from the WeChat official account “Internet of Things Think Tank” (ID: iot101). The author is Peng Zhao. It is published by 36Kr with authorization.