[International IP Briefing] China's Top Court: Free Gifts May Constitute Patent Sales

 

China’s Supreme People’s Court Holds That Free-of-Charge Provision of Infringing Patent Products May Constitute a “Sale”

On September 23, 2026, the Intellectual Property Court of the Supreme People’s Court of China published a case concerning the infringement of an invention patent, holding that providing an allegedly infringing patent product free of charge together with another product may constitute a “sale” under the Patent Law, depending on the substance of the transaction.

Overview

The Supreme People’s Court held that even where a promotional item is not separately priced, its provision may constitute a sale if it is offered on the condition that the customer purchases the main product and the economic value of the promotional item is effectively reflected in the profit from the sale of the main product. In such circumstances, the provision of the item forms part of the overall sales transaction.

Key Points

The key facts and findings of the Supreme People’s Court in Case No. (2024) Supreme People’s Court Intellectual Property Civil Final No. 1286 are summarized below.

Background of the Case

The patent holder owned an invention patent relating to a wire winding machine (“缠丝机”). The defendant sold three rebar cage welding machines and provided one wire winding machine free of charge with each machine. The wire winding machines were alleged to infringe the patent.

·         The wire winding machines were not separately priced in the purchase contract.

·         The defendant argued that because the machines were provided free of charge as promotional items, their provision did not constitute a “sale” under the Patent Law.

Whether the Provision Constituted a Sale

The Supreme People’s Court found that, in commercial transactions, providing a promotional item is a means of facilitating the sale of the main product. Once the purchaser pays for the main product, the economic value of the promotional item is effectively incorporated into the transaction.

·         Accordingly, the fact that the promotional item was not separately priced did not prevent the transaction from being considered a sale under the Patent Law, where the provision of the item was conditional on the purchase of the main product and its value was reflected in the seller’s sales profit.

Patent Infringement

The Court found that the wire winding machines manufactured and provided by the defendant incorporated all of the technical features set forth in Claims 1–3 of the patent and therefore fell within the scope of the patent claims.

Determination of the Number of Infringing Products Sold

The Supreme People’s Court relied, among other evidence, on VAT invoice records obtained from the tax authorities to determine the number of rebar cage welding machines sold by the defendant and used those figures as a basis for estimating the number of infringing wire winding machines sold.

·         Taking into account the defendant’s failure to provide relevant technical and sales information despite possessing such information, and its refusal to disclose the information on the grounds of commercial confidentiality, the Court determined that 468 wire winding machines had been sold between 2021 and 2024.

Calculation of Damages

The court of first instance awarded RMB 50,000 (approximately USD7,000) in statutory damages, including the patent holder’s economic losses and reasonable enforcement expenses. On appeal, however, the Supreme People’s Court recalculated the damages based on the infringer’s profits.

·         Based on the sales volume, average selling price, and operating profit margin, the Court estimated the defendant’s profits from the infringement at approximately RMB 547,700 (approximately USD76,000).

·         As the patent holder had claimed RMB 500,000 (approx. USD69,500) in damages, including reasonable enforcement expenses, the Supreme People’s Court awarded the full amount of RMB 500,000 (approximately USD69,500).

Final Judgment

The Supreme People’s Court upheld the first-instance order requiring the defendant to cease manufacturing and selling the infringing products, while increasing the damages from RMB 50,000 (approx. USD7,000) to RMB 500,000 (approx. USD69,500). The defendant’s appeal was dismissed.

The judgment is significant in clarifying that a product described as a “free gift” will not necessarily fall outside the scope of “sale” under China’s Patent Law. Where the gift is provided as part of a commercial transaction and its value is effectively incorporated into the price or profit of the main product, the transaction may be treated as a sale for patent infringement purposes.

  <Sources>

https://ipc.court.gov.cn/zh-cn/news/view-6092.html

https://ipc.court.gov.cn/zh-cn/news/view-6019.html

 

----------------------------------------------------------------- 

< Source of this post >

https://www.kiip.re.kr/board/trend/view.do?bd_gb=trend&bd_cd=1&bd_item=0&po_item_gb=&po_no=24748

 

A Hotel Run by Robots? We Checked in Through Patents

What if you checked into a hotel—and didn’t meet a single human employee during your entire stay?

A robot in the lobby turns to greet you. Another makes its way toward the guest rooms. A little later, your coffee arrives at the door on a robot’s tray. While you’re out, cleaning robots take care of the floors.

From a guest’s perspective, it all looks effortless.

From a robot’s perspective, almost none of it is.

Pressing an elevator button. Entering an elevator with people already inside. Carrying a cup of coffee without spilling it. Even figuring out what that mysterious thing on the floor actually is.

We barely think about these everyday actions. For a robot, each one is an engineering problem to solve. And some of those solutions have become patented technologies.

So this time, we brought together patents from several companies and checked into an imaginary robot-run hotel.

This isn’t a hotel where all of these technologies are actually operating under one roof. Think of it instead as a journey through patents—and a glimpse of what a stay in a future robot hotel might look like.

CHECK-IN — Is That Robot Looking at Me?

You walk into the lobby, suitcase in tow. As you approach the guide robot, it turns its gaze toward you. Move to the side, and its gaze shifts with you. For humans, this is almost automatic. We notice where the other person is and naturally turn our body and eyes toward them. But how does a robot know where to look?

A guide-robot patent currently listing Bear Robotics Korea as the applicant describes technology that uses information about nearby people to adjust the robot’s posture and gaze according to factors such as the direction of a person’s face. As that information changes during the interaction, the robot can use the updated information to determine what to do next.

In other words, the robot doesn’t simply stare in one direction and repeat the same routine. It can adjust its behavior according to the person it is interacting with.

The patent itself isn’t about processing hotel check-ins. But it offers an interesting glimpse into the kind of human-robot interaction that could make meeting a robot in a lobby feel a little more natural.

wipsglobal.com
KR10-2025-0024805 A
Guide robot and operation method of guide robot

Check-in complete. Time to head upstairs.

ON THE WAY UP — Before Finding the Way, “Where Am I?”

The robot heads toward the elevator. For us, finding our location can be as easy as opening a map and looking for the familiar blue dot. But inside a hotel, where GPS may not be something a robot can rely on, how does it figure out where it is?

A Pudu Robotics patent approaches the problem by using 2D LiDAR to measure the surrounding environment and comparing that information with previously constructed maps to estimate and refine the robot’s position and orientation.

In human terms, it’s a little like thinking: “I just passed the pillar next to the elevator, and there’s a long wall on my right... so I must be somewhere around here.” We use familiar sights as clues. A robot uses environmental data captured by its sensors and compares those clues with its maps.

Knowing the destination isn’t enough. Before deciding where to go next, the robot first needs to know where it is—and which way it is facing.

wipsglobal.com
US 12687637 B2
Fusion positioning method and apparatus, device and computer-readable storage medium

AT THE ELEVATOR — But Who Presses the Button?

The robot has made it to the elevator. And suddenly, a very ordinary problem appears. Who presses the button? For us, it takes one finger and about a second. For a robot, there is a little more to it.

A Pudu Robotics patent describes technology that uses sensors to identify the position of an elevator button panel and the target button, calculates the direction in which the button should be pressed, and then controls a robotic arm to press it. So the robot has to: find the panel → identify the right button → calculate its position and direction → move its arm → press.

Something we do without a second thought becomes a sequence of perception, calculation and physical action.

wipsglobal.com
CN 114348811 B
Robot, robot elevator taking method and device and storage medium

The button is pressed. The doors open. Problem solved? Not quite.

There may already be people, luggage or other obstacles inside. Another Pudu Robotics patent addresses what happens as a robot moves toward its position inside an elevator: it measures its distance from obstacles and determines whether to keep moving or stop based on that distance.

To us, pressing the button, looking ahead and stepping inside feels like one simple action. For a robot, pressing the button and safely entering the elevator are two separate engineering problems.

wipsglobal.com
US 12728534 B2
Method and system for controlling robot to take elevator

ROOM SERVICE — The Coffee Made It. So Did the Coffee.

You reach the room, unpack and order a coffee. A little later, a service robot arrives at your door. You pick up the cup from its tray. Which raises another question: Getting the coffee to your room is one thing. Getting it there still in the cup is another. A robot accelerates, turns and stops. Small bumps in the floor and sudden movements can all send vibrations through the tray.

A VD Robotics patent tackles this problem with a movable tray structure designed to absorb or reduce vibration, helping prevent drinks from shaking or spilling as the robot moves.

When we think about service robots, autonomous navigation and sensors usually get most of the attention. But if the mission is to deliver a drink safely, there’s engineering beneath the cup, too. To the guest, it’s just a cup of coffee. Underneath it is a small piece of engineering working hard to keep it that way. 

wipsglobal.com
KR 102830374 B1
Serving robot drink tray and robot including the same

WHILE YOU’RE OUT — Does Every Spot Really Need the Same Cleaning?

Coffee finished, you head out for the afternoon. Now it’s the cleaning robot’s turn. Would the best cleaning strategy simply be to cover every part of the floor once? Probably not.

Some areas may already be clean. Others may repeatedly get dirty.

One Samsung Electronics patent uses a pollution map containing location-specific contamination information to help identify a pollution source and control the robot’s movement based on its location.

Another Samsung patent uses an indoor map together with information such as dust levels and contamination to divide the cleaning area and plan different cleaning paths depending on conditions and cleaning modes. Put simply, the robot has two questions to answer: “Where is the dirt?” → “So where should I clean more carefully?”

Instead of treating every square meter the same, the robot can use information about the space and its condition to decide how to approach the job.

wipsglobal.com
KR 10-2025-0060869 A
Pollution source determination robot cleaner and operating method thereof

wipsglobal.com
KR 102616863 B1
Robotic vaccum cleaner and method for planning cleaning routes

What If the Robot Looks at the Floor and Thinks, “What is that?” Then the cleaning robot spots something on the floor. The camera sees it—but the robot isn’t quite sure what kind of contamination it is. Now what?

An LG Electronics patent offers a surprisingly interesting answer. When the robot cannot determine the type of contamination with sufficient confidence, it can apply an external stimulus—such as airflow, vibration or light—and use the resulting response to gather more information.

In other words, instead of simply giving up when looking isn’t enough, the robot can essentially say: “Let me poke at the problem a little and take another look.” When we talk about robot intelligence, we tend to imagine better cameras or more powerful AI. This patent takes a slightly different approach. If observing the world isn’t enough, interact with it—and learn something new from the response. That sounds surprisingly familiar.

wipsglobal.com
WO 2025244159 A1
Robot cleaner and control method thereof
(drawing texts translated by AI)

CHECK-OUT — The Hotel Looks a Little Different Now

The next morning, you check out and walk through the same lobby where your stay began. Yesterday, the guide robot simply seemed to turn and look at you. Now you can imagine the technology behind adjusting its gaze to a person’s direction.

At the elevator, you remember everything involved in pressing a button—and that even after the doors open, a robot still has to decide whether it is safe to move inside.

A room-service robot brings to mind the engineering hidden beneath a cup of coffee. A cleaning robot passing through the hallway makes you wonder what it sees on the floor, where it decides to clean next, and what it does when it isn’t quite sure what it is looking at.

The tiny things we barely notice in a hotel are anything but tiny for a robot.

Looking at a person.

Knowing where you are.

Pressing a button.

Moving safely around people.

Keeping coffee in the cup.

Deciding where more cleaning is needed.

Each is a small problem on its own. But solve enough of those small problems—and connect the solutions—and something much bigger begins to emerge: a hotel where living alongside robots feels completely ordinary.

Behind that ordinary experience, however, are more patented technologies than we might ever notice. So the next time you press an elevator button or pick up a cup of coffee at a hotel, consider a slightly different question: “How would a robot do this?” There may be a patent hiding behind the answer.

Serving robot (generated by AI)

The patents featured in this story were researched using WIPS CLIP. This article combines related technologies patented by different companies to create an imaginary hotel experience; it does not imply that these patents are implemented in any particular hotel or product.

 

----------------------------------------------------------

< Related articles and Sources>

World's first hotel entirely staffed by robots to open in 2027

Launching the World’s First Full-Scenario Robot-Serviced Hotel

 

Patent Data, Still Searching and Downloading One File at a Time? Get the Entire Database Instead

 

When conducting patent or trademark research, you probably search by keyword on a portal or search platform, download the results to Excel, and use them to prepare reports.

But what if you need to build an in-house system, train an AI model, or analyze millions of records at scale? Searching and downloading data every time quickly becomes inefficient.

That is where bulk IP data supply comes in.

WIPS provides large-scale IP databases that can be integrated directly into your own systems. Let’s take a closer look at what makes WIPS IP data different and how it can be used.

1. “Data for People” vs. “Data for Systems”

One of the most common questions is how purchasing IP data differs from simply downloading search results. The key difference is the purpose.

Search Result Downloads — Data for People

Users search for the information they need and download selected results to Excel for reports, research, or other business documents.

Data Products — Data for Systems

Large-scale patent, trademark, and design data covering more than 400 million records across 112 countries can be loaded directly into your own servers and systems.

Instead of being reviewed one record at a time, the data can power automated systems such as internal dashboards, analytics platforms, and AI models.

2. Why Not Just Use Free Public or Raw Data?

You may wonder, “Why not simply download free data from public databases or national IP offices?” The problem is that raw IP data is extremely difficult to use as-is.

Different formats across countries

Application numbers, publication numbers, company names, and other bibliographic information are represented differently from country to country and across languages. Standardizing them into a single database requires significant time, expertise, and cost.

Extensive processing required

Raw data does not necessarily provide analysis-ready information such as patent family and citation relationships, claim type classifications, or extracted technology keywords in a clean and structured form.

WIPS IP Data is preprocessed and standardized before delivery. Numbering systems from more than 22 countries are converted into a unified format with over 99.99% accuracy, while company names written differently across languages and jurisdictions are consolidated under standardized entity codes.

The result is analysis-ready data that can be used immediately for analytics, system integration, or AI training.

Beyond Patent Publications: More Than 500 Million Value-Added Data Records from WIPS

WIPS provides a wide range of enriched and structured data that cannot be found directly in patent publications.

· Patent Family & Citation Relationships

Applications filed in different jurisdictions are grouped by tracing priority relationships, while cited documents are linked by citation depth for more sophisticated relationship analysis.

· Claim Type Classification

Claims are classified into independent claims, dependent claims, and canceled/deleted claims.

· Keyword Extraction & Document Field Segmentation

Key technology terms are extracted from patent specifications, while detailed descriptions are structured into meaningful fields such as technical field, problem to be solved, solution, and effects.

· Patent Metrics & Indices

Nine types of patent metrics are available, including CPP (Citations per Patent), PFS (Patent Family Size), PII (Patent Impact Index), and TS (Technology Strength).

· AI-Based Patent-to-Industry Classification

Patents are mapped to manufacturing sectors under the Korean Standard Industrial Classification (KSIC), creating structured links between patent technologies and industries.

· Unified Company Codes

Different variations of company names are normalized and consolidated under a single company identifier.

gettyimagesbank.com

3. How Can Patent Data Be Leveraged in the AI Era?

More companies are now building their own AI, LLM, and RAG (Retrieval-Augmented Generation) systems. WIPS provides IP data products specifically designed to support these AI development environments.

AI-Optimized Summary Data

Key technologies, technical problems, solutions, and effects are summarized into concise information optimized for AI training and retrieval.

Structured Document Fields

Long and complex patent specifications are segmented into meaningful sections, helping AI models interpret and process patent documents more effectively.

AI- and Big Data-Ready Formats

Data can be delivered in developer-friendly formats such as JSONL and Parquet, making it easier to integrate directly into AI and big data workflows.

4. How Is WIPS IP Data Used in the Real World?

Here are some examples of how IP data can create practical business value.

1) R&D Strategy and Design-Around

Patent filings and changes in competitors’ patent rights can be integrated with internal R&D systems via API. Instead of repeatedly conducting manual searches, R&D teams can continuously access updated patent intelligence and use it to guide technology development and design-around strategies.

For example, automotive parts manufacturer COAVIS used WIPS patent data to support its R&D activities and filed 23 patents for core technologies.

2) New Product Development and Technology Scoring

Technology keywords and patent evaluation indicators—such as citation and market coverage metrics—can be analyzed to identify new business opportunities and objectively assess the value of technologies.

3) Dispute Prevention and Risk Management

Companies can monitor potentially infringing trademarks in overseas markets or analyze patterns in trial and appeal decisions to identify potential IP risks before they develop into disputes.

4) Corporate Investment and Evaluation

By combining corporate financial information with qualitative patent indicators, investors and analysts can gain a clearer picture of a company’s underlying technological capabilities and innovation potential.

5. How Is the Data Delivered?

WIPS IP data can be provided in a variety of ways depending on your company’s IT environment, including API integration and file-based delivery in formats such as XLSX, CSV, and JSON.

After receiving the initial complete dataset (Backfile), customers can receive only newly added or updated records (Current Data) on a weekly, monthly, or quarterly basis.

This makes it easy to maintain an up-to-date internal IP database without repeatedly rebuilding or downloading the entire dataset.

gettyimagesbank.com

Go beyond searching patents one at a time. Build your own IP intelligence infrastructure with WIPS data.

For inquiries about WIPS IP data products and bulk data supply, please contact global@wips.co.kr.