The United States Patent and Trademark Office (USPTO) director, Andrei Iancu, recently gave a talk at the IPO’s annual meeting, and he had some exciting news regarding subject matter eligibility, which affects the patentability of software. Ever since the Supreme Court Alice decision in 2014, many applicants have found the examination of software patents in the US inconsistent and unpredictable. Iancu readily acknowledged the existing problems with determining patentable subject matter, and in particular with determining whether or not protection is sought for an “abstract idea”, which is not patentable in the US under section 101. In an attempt to solve these problems the USPTO is now contemplating new guidance to simplify the test for an “abstract idea” and to improve the consistency and predictability of examination.

Iancu’s proposal is to make the question of subject matter eligibility one of “category”. Currently, US examiners tend to consider both inventiveness and sufficiency, when evaluating if the claimed subject matter is directed to an abstract idea. To avoid this muddling between subject matter eligibility on one hand and conditions of patentability on the other, Iancu proposes a simple two step test:

  1. Is the claimed matter a mathematical method, a business method or a mental process?
  2. If yes, is the matter integrated in a practical application?

If the answer to step 2 is "yes", then the claim is not an abstract idea, and relates to allowable subject matter.

Under current practice, if it is determined in step 1 that a claim is, for example, a mathematical method, then the next step would be to determine if the claim recites additional elements that amount to significantly more than the abstract idea (i.e. step 2 of the test from the Alice decision). This step is in itself so vague as to be impossible to determine with any degree of certainty. What is an “additional element”? When and how does it “amount to significantly more” than the abstract idea? These questions have led to further considerations, such as, are the recited elements well-known, routine and conventional? To determine if something is conventional it is necessary to look at the prior art, and all of a sudden the question of "abstractness" becomes confused with novelty and inventiveness.

Iancu’s proposed step 2, on the other hand, simply checks if the claimed mathematical method is "integrated in a practical application". This step looks much more clear, and a lot simpler than step 2 of Alice. Of course, we can ask what is meant by a “practical application”, but hopefully a common sense interpretation of this term will be sufficient to settle the issue.

The "practical application" test proposed by Iancu is also closer to the test applied by the European Patent Office (EPO), which requires the invention to be "technical". Inventions which are "technical" are generally also "practical", and vice versa. Therefore, if Icanu's proposal is accepted, it seems likely that the types of inventions which pass the US test will also pass the corresponding EPO test. This would lead to greater similarity of outcomes at the USPTO and EPO, and greater certainty for applicants.

Iancu finished his talk by noting that the new guidance will take some time to finalize, but also by stating his hope that the proposed guidance will help the USPTO out of their current rut when it comes to subject matter eligibility analysis.

Overall this looks like good news for applicants of software patents.