As part of the country’s efforts to strengthen intellectual property protection in newer areas of technology, China has focused efforts on broadening and clarifying the conditions under which patents may be issued for computer software. To these ends, in its 2023 Patent Examination Guidelines, the China National Intellectual Property Administration (CNIPA) added a special section on software patents which sets out clearer rules for the examination and offers illustrations to help in better understanding the software inventions that are more likely to be approved.
Overall, there is now a consensus in China that software can be patentable and that it should not be treated differently from other patentable subject matter. But in practice the eligibility of software-related inventions for patents will often be challenged where they are deemed only as rules and methods for “mental processes” and not as “technical solutions.” Strategies to overcome these challenges to patentability are discussed below.
Standards for Patentability
Article 2 of the PRC Patent Law provides for the definition of an “invention” for which patent rights may be granted, requiring any product or process eligible for patent protection to be a “technical solution.”
Article 5 of the Patent Law provides that: no patent rights shall be granted for any inventions that are contrary to the law or social morality, or that are detrimental to public interest; and no patent right shall be granted for any inventions where acquisition or use of the genetic resources, on which the development of the invention relies, is not consistent with the provisions of the law and administrative regulations.
Article 25 of the Patent Law excludes certain specific subject matter from patent protection, including rules and methods for “mental processes”.
In addition, article 20 of the Patent Law provides that the principle of good faith shall be followed in applying for patents and enforcing patent rights, and Rule 11 of the Implementing Regulations of the Patent Law further stipulates that any application shall be based on the actual inventive activities.
Software patents protect software-related inventions, which, as set forth in the Guidelines for Patent Examination, are inventions based in whole, or in part, on the flow of a computer program that can be executed by a computing device to control or process objects internal or external to the computing device.
Typically, claims of a software patent can be drafted into process claims or product claims. An example of a process claim is a follows:
A method for removing image noise, comprising:
- obtaining data of each pixel of an image to be processed that is inputted to a computer;
- calculating gray mean and gray variance of the image by using gray values of all pixels of the image; and
- removing the image noise.
Examples of product claims follow below:
- A computer apparatus, device or system, comprising a memory, a processor and a computer program stored on the memory, characterised in that the computer program is executed by the processor to perform the steps of the method as in process claims.
- A computer-readable storage medium on which a computer program or instructions are stored, characterised in that the steps as in the process claims are performed when the computer program or instructions are executed by a processor.
- A computer program product comprising a computer program or instructions, characterised in that the steps as in the process claims are performed when the computer program or instructions are executed by a processor.
- Apparatus for removing image noise, comprising:
- means for obtaining data of each pixel of an image to be processed that is inputted to a computer;
- means for calculating gray mean and gray variance of the image by using gray values of all pixels of the image; and
- means for removing the image noise.
Examination of subject matter eligibility of software patents
Software patent claims typically include features based on the steps of a computer program flowchart. In some cases, the claims include features of both hardware and software. In other cases, the claims are drafted based entirely on software, especially for inventions relating to business methods or improvements to algorithms.
Due to the characteristics of software patent claims, the examination of subject matter eligibility focuses on evaluating whether the claimed subject matter falls under rules and methods for mental processes under Article 25 of the Patent Law, and whether the claims constitute technical solutions according to the provisions of Article 2 of the Patent Law.
Rules and methods for mental processes
Rules and methods for mental processes refer processes according to which people think, express ideas, make judgments and recall via their memory.
If a claim only relates to rules and methods for mental processes, then it is ineligible to be granted a patent right. If, except for the title, all limitations of a claim are rules and methods for mental processes, then the proposed invention thus essentially only relates to rules and methods for mental processes and shall not be granted a patent right. For example, methods and systems of managing organisations, traffic rules, operating instructions for an instrument or apparatus, rules and methods of various games or entertainment, food recipes, surveys about diseases and population censuses are generally rules and methods for mental processes, and are therefore patent ineligible.
A claim including both rules and methods for mental processes and technical features, considered as a whole, is not necessarily considered to constitute purely rules and methods for mental processes, and thus shall not be excluded from patent protection. For example, a claim drawn to a business method patent shall not be rejected as rules and methods for mental processes if it includes both a business method and technical features, such as involving computers, servers and communication networks.
Technical solution
Under the Guidelines for Patent Examination (2023), a “technical solution” is a set of technical means following the laws of nature to solve a technical problem, and technical means are embodied by technical features. There are three elements that constitute a technical solution: (i) to solve a technical problem, (ii) to use technical means following the laws of nature, and (iii) to achieve technical effects accordingly.
In some cases, it may be controversial to determine whether a software-related invention is a technical solution or not. This is due to various concepts involved, such as the “laws of nature” and the term “technical”, lacking clear and precise definitions. This is further complicated because their connotations may even change with the development of human knowledge and perception.
To unify the criteria for identifying a technical solution, the Guidelines for Patent Examination (2023) provide some typical examples of technical solutions as follows.
- An industrial process control program that is executed to control various phases of the industrial process. For example, a method for controlling the rubber moulding process by using a computer program, and a method for measuring the viscosity of a liquid by using a computer program.
- A data processing program that is executed to process external technical data. For example, a method for removing image noise.
- A program that is executed to improve internal performance of the computer. For example, a method for expanding the storage capacity of a mobile computing device.
- A program that is used for processing big data in a specific application field.
AI, big data, business methods, and Internet-based applications
Claims of solutions in these areas involve features of algorithms or features of business methods. Just a few years ago, it was very difficult to persuade examiners that such claims were technical solutions and should not be excluded from patent protection.
Based on economic and technology developments and the continuous efforts of applicants, China’s central government decided to strengthen patent protection for inventions in these areas, and, in intellectual property policies issue in recent years, has referred to these these areas as “new fields” and “new business forms”.
Accordingly, the CNIPA amended the Guidelines for Patent Examination several times, and added a new special section governing the examination of inventions in these areas).
In the section, it is emphasised that features of an algorithm or features of business methods cannot simply be separated from technical features recited in a claim during examination, but rather they should be evaluated comprehensively as a whole.
For example, a method for analysing propensities to use e-coupons comprises the following steps:
- classifying the e-coupons according to information on e-coupons to obtain types of the e-coupons;
- obtaining user sample data according to the application scenarios of the e-coupons;
- extracting user behaviour characteristics from the user sample data based on user behaviours, wherein the user behaviours include browsing webpages, keywords searching, following, adding to cart, purchasing and using e-coupons;
- training an e-coupon propensity recognition model for a different type of e-coupons by using the user sample data as training samples and the user behaviour characteristics as attribute tags; and
- obtaining propensities to use different types of e-coupons by predicting probabilities of using the e-coupons with the trained e-coupon propensity recognition models.
This claim relates to the processing big data for e-coupons. The correlation between user behaviours and propensities to use e-coupons are objective and follow the laws of nature. Therefore, the invention qualifies as a technical solution.
Consider further, a method for training a deep neural network model, comprising the following steps:
- when the size of training data is changed, calculating the training time for each of the training scheme candidates on the changed size of the training data;
- choosing, from the training scheme candidates, the one for which the training time is shortest as the optimal training scheme for the changed size of the training data; the training scheme candidates include a single-processor training scheme and a multi-processor training scheme based on parallel data processing; and
- training the model on the changed size of the training data by using the optimal training scheme.
This claim relates to training a deep neural network model. The training method involves choosing between a single-processor training scheme and a multi-processor training scheme. That is, the invention has special technical correlation to the internal structure of the computing system, and improves the performance of the hardware. Therefore, the invention provides a technical solution.
Tips for overcoming rejections for ineligible subject matter
Claims
Generally, it is suggested to draft the claims, especially the first independent claim, with proper scope and with some well-recognised technical features. If an invention references traditional industrial areas, is used to process technical data or it can improve the performance of the computing system, the best practice is to include relevant features in the claims for the examiner. If an invention relates to the new fields and new business forms, the best practice is to include features that limit the solution to a particular application referred to in the application and to include features that constitute the relevant hardware environments.
Overly general independent claims, especially the first independent claim, should be avoided. Words purely for describing business or finance should not be included in claims – for example, “revenue”, “profit”, “game” and “bid”.
Description
If an invention references traditional industrial areas, is used to process technical data or it can improve performance of the computing system, it is sufficient to simply describe the steps of the software and the achieved technical effects. If an invention relates to new fields and new business forms, the description should sufficiently describe steps of the software and its operational environment, and describe achieved effects from various aspects, especially objective non-business effects – for example, improvements to user experience, increasing efficiency and/or increasing accuracy.
The description should not overemphasise non-technical aspects, especially problems and achieved effects.
Response to office actions
If an ineligible subject matter rejection is received, one good strategy is to try to amend the claims as suggested above and state in the response that the solution follows objective laws (i.e., principles that are objective, independent of human will or subjective goals), and the solved problem and achieved effects are also objective.
For an invention in new application fields, sometimes it is a tough task to determine whether the invention follows the laws of nature, or whether it is technical or not. The Guidelines for Patent Examination (2023) give many examples and accompanying analysis on why some qualify as technical solutions and why others do not. But the reasoning underlying some examples is not clear or persuasive enough. In responding to the rejection for ineligible subject matter, applicants may compare the rejected invention with the typical examples listed in the Guidelines for Patent Examination (2023), and point out the similarities between the rejected invention and the allowable examples under the Guidelines to the examiner.
Conclusion
China is now more open to patent protection for inventions in new technologies. The CNIPA may modify the rules from time to time as appropriate to adapt to progress in technology, changes in society and developments in the economy. The Supreme Court of China has issued some guiding cases relating to software patents and may issue other guiding cases or judicial guidelines relating to software patents. Applicants should consult local counsel in a timely manner and keep a close eye on changes of China’s policies and practices.