Excluded subject matter – a reminder of the importance of applying the statutory test

Facts

Mr Reaux-Savonte applied for a patent, published as GB 2546236, entitled "The genome and self-evolution of AI". The examiner was of the view that there was a lack of technical detail in the application and that the invention was nothing more than data structured in a modular, hierarchical and self-contained manner, and therefore was excluded under section 1(2)(c) of the Patents Act as a program for a computer. The Hearing Office was of the same view and, in his decision, refused the application. Mr Reaux-Savonte appealed to the High Court, his central contention being that the Hearing Officer applied the law relating to the exclusion from patentability in an overly restrictive manner. Daniel Alexander QC, sitting as a deputy judge, dismissed the appeal.

Approach

This was a slightly unusual case in that that it involved only one objection to patentability - the computer program exception. Mr Alexander noted that cases usually involve the interplay between at least two exclusions, e.g. a mathematical method and/or a program for a computer and/or a business method. This did not mean that a single objection was impermissible, but he needed to proceed with appropriate care.

Mr Alexander also discussed how the contribution or alleged contribution in the patent application should be identified, and the extent to which appellate Courts should take it upon themselves to evaluate the primary facts. He accepted that what an applicant said about the technical contribution should be taken into account but noted that the "primary point of reference" for the technical contribution had to be the patent application document. As to considering this matter on appeal, Mr Alexander recorded that the appellate Court should not interfere with the Hearing Officer's assessment unless there was an error of principle or the evaluation was "sufficiently clearly erroneous".

The decision

The Hearing Officer had identified the invention's contribution to the art as being a particular way of structuring and organising data that may allow for the production and evolution of future AI code, the word "genome" being used in the application purely by way of analogy to refer to characteristics of a particular kind of computer code.

Having assessed the contribution, the Hearing Officer turned to consider whether the contribution was technical. In finding that it was not, Mr Alexander held that he correctly applied the signposts identified in AT&T Knowledge Ventures/CVON Innovations v Comptroller General of Patents [2009] EWHC 343 (Pat) and reformulated in HTC v Apple [2013] EWCA 451. Only the first three signposts were relevant.

First signpost: whether the claimed technical effect has a technical effect on a process which is carried on outside the computer

Mr Reaux-Savonte suggested that the invention enabled the healing and reproduction of AI genomes (i.e. computer code) in an analogous way to biological genes in nature, thereby allowing the creation of physical systems outside a computer. However, Mr Alexander held that such a potentially purely theoretical impact which was not shown to be contributed by the invention could not satisfy the first signpost.

Second signpost: whether the claimed technical effect operates at the level of the architecture of the computer

Mr Alexander agreed with the Hearing Office that, since the invention intrinsically related to the organisation and structuring of the data being processed which was not at the architecture level, the second signpost was not satisfied.

Third signpost: whether the claimed technical effect results in the computer being made to operate in a new way

The Hearing Officer concluded that he was unable to find anything in the application that suggested that a computer system was being made to operate in a new way and noted that a computer system operating on new code did not imply that the system worked in any way differently to how it would with the old code. Mr Alexander agreed.

Discussion

Cases relating to the exclusions to patentability, particularly where the invention is a program for a computer, have caused real difficulties for the Courts. Lewison LJ in HTC v Apple, articulated his frustration when he said that:

"It is, to me at least, regrettable that because these apparently simple words have no clear meaning both our courts and the Technical Boards of Appeal at the EPO have stopped even trying to understand them. However, we are so far down that road that "returning were as tedious as go o'er". Instead, we are now engaged on a search for a "technical contribution" or a "technical effect". Instead of arguing about what the legislation means, we argue about what the gloss means. We do not even know whether these substitute phrases mean the same thing …"

Since then, the law has settled down and the Manual of Patent Office Practice gives clear guidance as to how to apply the law relating to section 1(2). However, it is just that – guidance. The Manual does not have the force of law. Therefore, Mr Reaux-Savonte was not going to succeed in his submission that the Hearing Officer had applied the guidance in the Manual in an overly restrictive manner (particularly the passages explaining the signposts and the passage relating to the patentability of AI). Instead, he needed to show that the Hearing Officer had made an error or principle of clearly misapplied the statutory test which he had not done.

Mr Alexander used the opportunity to point out that there was limited value in an examination of the Manual's gloss which was itself a gloss on the statutory test. Analysis of the signposts had been repeatedly stated only to provide indicia, none of them being individually conclusive as to whether the statutory test was satisfied. The key take-home message was that it was important not to lose sight of the statutory test or the invention as disclosed.