16 October 2020

How does the examination of AI and blockchain inventions in China compare to that at the European Patent Office?

On the last day of 2019, the China National Intellectual Property Administration (CNIPA) announced the latest amended guidelines for patent examination effective on 1 February 2020. This amendment was announced only three months after the previous amendment made in September 2019. This unusually speedy amendment to the guidelines reflects the Chinese government’s plans to strengthen the protection of IP rights in the fast-growing high-tech fields of artificial intelligence (AI) and blockchain.

What’s new

A new section (Part II, Chapter Nine, Section Six, ‘the examination guidelines for inventions that contain algorithms or business rules and methods’) has been added to the guidelines. This new section aims to clarify the examination of inventions relating to AI, internet plus[1], big data and blockchain, which normally contain intellectual activities such as algorithms or business rules and methods.

The new guidelines indicate that the ‘as-a-whole’ principle for considering patentable subject matter (Article 25.1(2) and Article 2.2), as well as assessing novelty and inventive step – technical features and features relating to algorithms or business rules and methods – should not be separated in a claim. Whether a claim constitutes a technical solution should be assessed based on the technical means, the technical problem and the technical effect of the claim as a whole.

The ‘interaction-consideration’ principle for assessing inventive step is as follows: to be considered as contributing to the inventive step of a claim, features relating to algorithms or business rules and methods should be closely integrated with technical features (‘mutual supportive and interactive relations in function’) to form a technical means to solve a technical problem and result in a corresponding technical effect.  Otherwise, those features relating to algorithms (in effect mathematical methods) and business rules or methods would be considered non-technical and therefore not relevant when assessing the inventive step of a claim, in a manner similar to that taken in Europe.

EPO and CNIPA new guidelines

Details about the new section of the Chinese guidelines can be found in Table 1 (emphasis added) with a brief comparison between the European Patent Office (EPO) and the CNIPA guidelines.

EPOCNIPA (Part II, Chapter 9, Section 6x)
Case Law of the Boards of Appeal of the EPO, I.A.2.2.2  
“The exclusion applies if a claim is directed to a purely abstract mathematical method and the claim does not require any technical means”.
6.1.1 Examination using Article 25.1(2)  
“If a claim relates to an abstract algorithm or a pure business rule and method, and does not contain any technical features, the claim is excluded as a rule and a method of intellectual activity under Article 25.1(2) and shall not be granted.”
Guidelines G-II 3.3  
“If a claim is directed either to a method involving the use of technical means (e.g. a computer) or to a device, its subject-matter has a technical character as a whole and is thus not excluded from patentability under Art. 52(2) and (3)“.

Case Law of the Boards of Appeal I.D.9.1.1  
“In order to be patentable, the subject-matter claimed must therefore have a “technical character” or to be more precise – involve a “technical teaching”, ie an instruction addressed to a skilled person as to how to solve a particular technical problem using particular technical means.”  
6.1.2 Examination using Article 2.2    
“If a claim for protection as a whole is not excluded under Article 25.1(2) of the Patent Law, then it is necessary to examine whether it is a technical solution specified in Article 2.2 of the Patent Law. When examining whether a claim containing algorithm or business rule and method features constitutes a technical solution, it is necessary to consider all the features in the claim as a whole. If the technical means recited in the claim uses the laws of nature to solve a technical problem, and thus obtains a technical effect in conformity with the laws of nature, the solution recited in the claim is a technical solution specified in Article 2.2 of the Patent Law.”  
GL G-VII 5.4  
“When assessing the inventive step of such a mixed-type invention, all those features which contribute to the technical character of the invention are taken into account.   These also include the features which, when taken in isolation, are non-technical, but do, in the context of the invention, contribute to producing a technical effect serving a technical purpose, thereby contributing to the technical character of the invention. However, features which do not contribute to the technical character of the invention cannot support the presence of an inventive step (T 641/00). Such a situation may arise, for instance, if a feature contributes only to the solution of a non-technical problem, e.g. a problem in a field excluded from patentability (see G‑II, 3 and sub-sections).”
6.1.3 Examination on novelty and inventive step  
“In the examination of novelty of an invention patent application containing algorithm or business rule and method features, all the features recited in a claim shall be taken into account, and here all the features include both technical features and algorithm or business rule and method features.”  

“In the examination of inventive step of an invention patent application that contains both technical features and algorithm or business rule and method features, the technical features should be considered as a whole with the algorithm or business rule and method features that have mutual supportive and interactive relations in function with the technical features. “Mutual supportive and interactive relations in function” means that algorithm or business rule and method features are closely integrated with technical features to form a technical means to solve a technical problem and can obtain a corresponding technical effect.

For example, if the algorithm in a claim is applied to a specific technical field and can solve a specific technical problem, then it can be considered that the algorithm features and technical features have mutual supportive and interactive relations in function, and the algorithm features become a part of the adopted technical means, in the examination of inventive step the contribution of the algorithm features to the technical solution should be considered.

In another example, if the implementation of business rule and method features in a claim requires an adjustment or an improvement of technical means, then business rule and method features and technical features could be considered to have mutual supportive and interactive relations in function, and in the examination of inventive step the contribution of business rule and method features to the technical solution should be considered.
Table 1. EPO and CNIPA new guidelines

CNIPA examples

Section 6.2 provides 10 examples illustrating the application of the new guidelines in various scenarios from both positive and negative aspects. Examples 1 to 6 demonstrate whether the subject matter in the claimed invention is patentable. Example 1 relates to a method for establishing an abstract mathematical model, which should be excluded from patentability under Article 25.1(2). Examples 2 to 6 discuss what constitutes a technical solution in the fields of AI, business models and blockchain specified in Article 2.2 of the guidelines.

For instance, example 4 describes a method for preventing blockchain business nodes from leaking user privacy data in the alliance chain network. This claimed invention addresses a technical problem of providing more secure blockchain data – business nodes only establish connections to limited objects, by carrying the Certificate Authority (CA) certificate in the communication request and configuring the CA trust list in advance to determine whether to establish a connection. Such technical means achieve the technical effect of securing communications between business nodes and reducing the possibility of business nodes leaking private data. Therefore, the solution in this example is a technical solution specified in Article 2.2.

In contrast, example 5 relates to a method for promoting users’ consumption. This method, however, does not constitute a technical problem because a computer is only executed to determine the rebate amount based on the user’s consumption amount according to the specified rules. Therefore, in this application, no technical means are used, and the only effect of this claimed invention is to promote users’ consumption. The subject-matter in this patent application thus is excluded from patentability.

Examples 7 to 10 demonstrate whether the subject matter in patent applications involves an inventive step, which will be compared with EPO’s examples in a later section.

SectionTitle
6.1.1 Article 25.1(2)
An abstract algorithm or a pure business rule and method?
Ex. 1 – A method for establishing a mathematical model
6.1.2 Article 2.2  
A technical solution?
Ex. 2 – A method for training a convolutional neural network model

Ex. 3 – A method for using shared bicycles

Ex. 4 – A communication method and device between blockchain nodes

Ex. 5 – A method of consumption rebate

Ex. 6 – A method for analysing an economic sentiment index based on electricity consumption characteristics
6.1.3
Novelty and inventive step
Ex. 7 – A method for detecting the falling state of a humanoid robot based on multi-sensor information

Ex. 9 – A method of logistics distribution

Ex. 8 – A multi-robot path planning system based on a cooperative co-evolution and multi-group genetic algorithm

Ex. 10 – A method for visualising the evolution of dynamic viewpoints
Table 2. Examples provided in the new guidelines

EPO v CNIPA – patentable subject matter

For the purposes of comparison, these new guidelines (section 6.1.1 and section 6.1.2) relating to patentable subject matter appear to be close to the EPO’s approach (as compared in the following Figure 1 and in the previous Table 1).

Figure 1. Patentable subject matter examination

EPO v CNIPA – inventive step

The EPO guidelines (GL G-VII 5.4) and the case law (T 641/00) distinguish between three groups of features:

i) technical features,

ii) non-technical features, and

iii) features which, when taken in isolation, are non-technical, but do, in the context of the invention, contribute to producing a technical effect serving a technical purpose, thereby contributing to the technical character of the invention.

When following the approach set out in the new Chinese guidelines for assessing inventive step (section 6.1.3), algorithm or business rule and method features may be equivalent to the features in the EPO’s group iii. It might be worth noting that the CNIPA currently only considers these two particular kinds of ‘non-technical’ features (i.e. algorithms or business rule and method features) to be able to contribute to inventive step. The CNIPA’s approach seemingly first considers whether algorithm or business rule and method features and more traditional ‘technical features’ interact with each other functionally in the context of the invention to solve a technical problem and result in a corresponding technical effect. If yes, then the contribution of algorithm or business rule and method features should be considered.

EPO examples (inventive step)
5.4.2.1 Example 1 – Method of facilitating shopping on a mobile device
5.4.2.2 Example 2 – A computer-implemented method for brokering offers and demands in the field of transporting freight
5.4.2.3 Example 3 – A system for the transmission of a broadcast media channel to a remote client over a data connection
5.4.2.4 Example 4 – A computer-implemented method for the numerical simulation of the performance of an electronic circuit subject to 1/f noise
Table 3. Examples provided in the guidelines G‑VII, 5.4 (‘Claims comprising technical and non-technical features’)

Two CNIPA examples (example 7* and example 10** detailed in post-script) and two EPO examples (5.4.2.2 example 2 and 5.4.2.4 example 4) are selected to demonstrate whether technical features and algorithm or business rule and method features interact with each other functionally in the context of the invention.

In example 7, it is considered that algorithm features – i.e. the fuzzy algorithm in step (2) – are closely integrated with technical features (i.e. the robot information in step (1)), because the input parameters of the fuzzy algorithm are limited to the robot information to determine the stability state of the humanoid robot. In one EPO example (5.4.2.4 example 4), ‘the mathematically expressed claim features’ when considered in isolation, represent a mathematical method with no technical character. However, these features contribute to the technical character of the method because the claim is limited to a computer-implemented method in which this mathematical method serves a technical purpose of ‘circuit simulation’. In other words, the ‘mathematically expressed claim features’ and the technical features of ‘circuit simulation’ can be considered to interact with each other functionally in the context of the invention.

In example 10, the CNIPA explains that the technical means for visualisation could be used with different sentiment categorisation rules (the algorithm feature). Therefore, the sentiment categorisation rule and the visualisation method do not interact with each other functionally. Similarly, the EPO provides an example in its guidelines (5.4.2.2 example 2) in which the features defining a business method were ‘easily separable’ from the technical features of its computer implementation. Thus, the business method features and technical features cannot be considered to interact with each other functionally in the context of the invention.

Conclusion

To conclude, the new CNIPA guidelines bring some welcome clarity to what was a grey area of determining the patentability of an application involving business rules and methods, algorithms and mathematical methods. Previously, it was very common to see Chinese Examiners assessing inventive step based on divided features in a claim, and there was no clear guidance on how to examine ‘non-technical’ features. The new guidelines, therefore, tend to be more friendly to patent applicants because of the new ‘as-a-whole’ principle and the ‘interaction-consideration’ principle. This fast-track amendment indicates that, in order to keep up with the times, the CNIPA is maintaining a positive attitude towards AI and blockchain-related inventions.

Post-script

Example 7 *

Claim 1 recites

“A method for detecting the falling state of a humanoid robot based on multi-sensor information, which is characterized by the following steps:

(1) fusing attitude sensor information, zero-moment point ZMP sensor information and robot walking stage information to establish a sensor information fusion model with a layered structure;

(2) using the front and rear fuzzy decision system and the left and right fuzzy decision system to determine the stability of the robot in the front and rear direction, and the left and right direction respectively, the specific steps are as follows:

① determining the walking stage of the robot according to the contact between the robot support foot and the ground and offline gait planning;

②applying fuzzy inference algorithm on the position information of ZMP;

③applying fuzzy inference algorithm on the pitch or roll angle of the robot;

④ determining the output membership function;

⑤ determining fuzzy inference rules according to steps ①~④;

⑥ de-fuzzifying.“

D1 is considered as the closest art and discloses step (1) of the method. Therefore, the difference between D1 lies in step (2) – i.e. implementation of the fuzzy decision algorithm. Compared to D1, the technical problem solved by the claimed invention is how to determine the stability state of the robot and accurately predict its possible fall direction. The altitude information, ZMP (Zero Moment Point) position information and walking stage information in step (1) are used as input parameters, and the fuzzy algorithm in step (2) outputs information to determine the stability state of the humanoid robot, which provides a basis for further accurate posture adjustment instructions. Therefore, the algorithm features and technical features have mutual supportive and interactive relations in function, and they are closely integrated and jointly constitute a technical means to solve a technical problem and thus obtain a corresponding technical effect.

Example 10 **

Claim 1 recites

“A method for visualising the evolution of dynamic viewpoints, comprising:

Step 1 – a computing device determines a sentiment membership and sentiment category of the information in the collected information set, the sentiment membership of the information indicates how likely the information belongs to a sentiment category;

Step 2 – the sentiment category is positive, neutral, or negative, the specific categorisation method is: If the value r of the number of likes p divided by the number of dislikes q is greater than the threshold value a, then the sentiment category is considered as positive, if the value r is less than the threshold b, then the sentiment category is considered as negative, if b≤r≤a, then the sentiment category is neutral, wherein a>b;

Step 3 – based on the sentiment category of the information, the geometric layout of the sentiment visualisation of the information set is automatically established, the horizontal axis represents the time of information generation, and the vertical axis represents the amount of information belonging to each sentiment category;

Step 4 – the computing device colours the established geometric layout based on the sentiment membership of the information, and colours the information on each sentiment category layer according to the gradual order of the information colour.”

D1 discloses a sentiment-based visualisation analysis method, where time is represented as a horizontal axis, the width of each colour band represents a measure of sentiment at that time, and different colour bands represent different sentiment.

The difference between the solution of the claimed invention and the closest prior art D1 is the new sentiment categorisation rule proposed in step 2. It should be noted that the same technical means for colouring could be used with different sentiment categorisation rules. Therefore, the sentiment categorisation rule and the visualisation method do not have mutual supportive and interactive relations in function.

In addition, the problem of ‘visualisation’ has been solved in D1. Compared to D1, the claimed invention only proposes a new rule for sentiment categorisation, which does not actually solve any technical problem or make a technical contribution to the art. Consequently, the claimed technical solution of the claimed invention does not involve an inventive step in light of D1.


[1] Similar to Information Superhighway and Industry 4.0, the Chinese government has created its own “Internet Plus” initiative to transform, modernize and equip traditional industries to join the modern economy.