8.1 Information modelling
Information modelling, though a precursor for program design, has been considered to be a non-technical activity as such. It might contribute to the technical character only if specifically applied in a technical environment.
8.2 Database technology
Database technology, in terms of the technical functions and data structures actually stored in the computer, has been found to be technical.
Likewise, the exchange of data between different application programs using functional data structures (e.g. clipboard formats) were considered an extension of the the inner operation of a computer system and were thus found to be patentable. In addition, programs that refer to garbage collection and aspects of data retrieval were also considered to provide a technical effect.
8.3 Mathematical methods/simulation
On the one hand, mathematical methods as such are also on the list of non-inventions. On the other hand, mathematics easily qualifies for technical applications. In this respect, it was found that a linear combination for automatically selecting a database management system in a data consistency management system has technical character, since it significantly contributes to the operation of the system.
Even a simulation that is essentially based on mathematical models can solve a technical problem if it produces a technical effect that exceeds the mere implementation of the simulation.
8.4 Business methods/Financial transaction
Such concepts - that are excluded as such - are hardly suitable for technical applications and, thus, must not be considered when inventive step is assessed.
What could, however, turn out to be patentable are special implementation aspects using hardware designs or program constructs that, in themselves, have a technical character. It is therefore important to include as many concrete technical implementation details as possible to support sufficiency of disclosure and to increase the chances of obtaining a patent for innovations in this field.
8.5 Information/Translation
Pure information contents are not patentable. However, the use of a piece of information in a technical system, or its usability for this purpose, may confer a technical character on the information itself in that it reflects the properties of the technical system in which it exists, e.g. by being specifically formatted and/or processed. Linguistic aspects of a translation process may also generally assume a technical character if they are used in a computer system and form part of a solution to a technical problem.
8.6 Graphical user interfaces (GUIs)
Judicial practice is reluctant to attribute technical character to the design of graphical user interfaces, particularly if they are only based on aesthetic considerations or solely aim to facilitate human perception or mental processing. Visual indications of the internal states of a technical system in the form of visual feedback for human interaction with the system have, however, been accepted as technical. All in all, for the time being it appears that, even though they use one and the same structured approach to assessing patentability, different Boards of Appeal are not consistent in drawing the line in respect of technical character of GUIs. Rather, the assessment seems to depend on whether a broader or narrower construction of the meaning of “presentation of information” (which are excluded “as such”) is applied.
8.7 Computer games
Computer games naturally involve schemes, rules and methods for playing games, software and presentations of information through graphical user interfaces. All of these aspects have to be carefully examined to see whether they make a technical contribution. Aspects purely driven by game rules have to be ignored.
8.8 Bioinformatics
Albeit not abundant at present, the existing case law throws light on the realms of technicality in bioinformatics and follows the established view in other technical fields that features excluded “as such” must not be ignored or separated if they serve a technical purpose and thus contribute to the technical character of the claimed subject-matter. In particular, an automated genotype determination is technical, and improving the confidence of the genotype estimate relates to a technical problem.
Any means contributing to the solution of that problem therefore qualifies as technical means.
8.9 “Big data” and artificial intelligence
Not least due to the rapid development of the Internet and the success of smartphones, the global data pool has grown almost exponentially in the last few years. The development of modern high-performance processors and the steadily growing storage media make the efficient analysis of “big data” possible. Knowledge obtained from this in conjunction with artificial intelligence (AI) has created voice and face recognition systems, autonomously driving cars as well as adaptive production facilities. By now, the latter play a central role in the area of Industry 4.0, with the use of AI not only improving known manufacturing processes but also the automation of drafting and design processes. But also in other high-tech sectors, such as medical technology and the pharmaceutical industry, AI systems are now increasingly used. The core of AI systems is usually constituted of software that controls and monitors the training of self-learning AI systems.
As regards protection of AI-related inventions, a variety of aspects arise which have to be taken into consideration in the process of drafting and granting patents. What is paramount here is the protection of AI systems as such, of “big data” used for training the neural networks used in AI systems and of products manufactured by such systems.
For software-implemented components of AI systems, the above-described rules and approaches from the field of computer-implemented inventions are used for determining patentability. The focus is thus also on assessing whether individual features of the invention solve a technical problem by technical means.
Since the software-implemented functioning of an AI system is often not or hardly predictable, especially the correct explanation of the functioning of an AI system is a great challenge in drafting a corresponding patent application as the latter has to disclose the invention in a manner sufficiently clear and complete for the person skilled in the art to be able to carry it out (or to rework it). Therefore, all components of an AI-related invention should comprehensively be acknowledged in an associated patent application, from the training data, the used traning method, and the configuration of the neural network underlying the AI system to the manufactured product or result. Ideally, the patent application also contains measurement data that make the functioning of the AI system underlying the invention plausible.
The protection of products generated by an AI system is less problematic, because here the conventional standard of examination for assessing inventive step applies.