Skip to main content

I was fortunate enough to attend a presentation by the Honourable Justice Jagot at the 2022 IPSANZ Conference which was subsequently published as an article in the Intellectual Property Forum Journal.1 Her Honour opined on how evidence relating to lack of inventive step is gathered for the purposes of court proceedings, and how the desire to avoid hindsight in the gathering of this evidence has led to the development, by patent attorneys and lawyers, of rituals which are time consuming, expensive and may not actually achieve the objective of avoiding hindsight.

For some time, I have been puzzled by a tangential but related point – on what basis do patent attorneys and lawyers decide who the best expert(s) will be to represent the notional person skilled in the art (“PSA”) in a particular case?

What does a PSA do?
Let’s take a step back. We are familiar with the concept of a “reasonable person” or “the man on the Clapham omnibus” in tort, criminal and contract law. “Reasonable person” is defined by the Merriam-Webster dictionary as:2

A fictional person with an ordinary degree of reason, prudence, care, foresight, or intelligence whose conduct, conclusion, or expectation in relation to a particular circumstance or fact is used as an objective standard by which to measure or determine something (as the existence of negligence).

In Australian patent law, we have the concept of a PSA. The PSA is a hypothetical construct that is not that different from the “reasonable person” – they too may be said to possess ordinary skills and intelligence, and their expectation in relation to a particular set of information, and what can be routinely done with it, is used as an objective standard to assess the patentability of an invention.

Specifically in the context of Australian patent law, the PSA is a skilled but non-inventive worker in the relevant field of technology, taken to be informed of the common general knowledge in the art. Their expertise may vary from that of a tradesperson to that of a scientist (depending on the nature of the invention). The PSA is assumed to have a practical interest in the claimed invention and want to put the claimed invention to use.3 Depending on the complexity of the relevant field, the common general knowledge of the PSA may be quite sophisticated and may even be derived from multiple fields. In the case of the latter, the PSA may be a team of people with different types of expertise.

The PSA is a feature of many aspects of Australian patent law. I will borrow some of Justice Jagot’s introductory statements from her article, which highlight the way in which a PSA is invoked in patent law:4

The “art” is the subject-matter or field of the patent: Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183 at 235. … Identifying the field or fields of the patent is important because the person skilled in the art is the person with expertise in that area who is to be attributed with the common general knowledge of the relevant field or fields of the patent: JMVB Enterprises Pty Ltd v Camoflag Pty Ltd [2005] FCA 1474; (2005) 67 IPR 68 at [91].

So, we get to the relevant PSA by determining the field or fields of the patent. Patents are also to be construed through the eyes of the PSA. As are prior art documents. Therefore, the PSA is used to interpret the meanings of terms in patent claims, and to help us understand what prior art documents disclose or “teach”, and what a patent specification enables a PSA to do. Thus, the PSA is invoked when assessing the grounds of novelty, support and sufficiency, and in determining claim scope.

The PSA also plays a central role in determining whether an invention is obvious, or “lacks an inventive step”. As noted above, the PSA is imbued with “common general knowledge”, which is the knowledge that has been assimilated and accepted by the majority of those in the relevant field. It includes material and information to which a PSA would have regard to as a matter of course. An invention is taken to involve an inventive step, unless it would have been obvious to a PSA in light of the common general knowledge and any prior art information that was publicly available before the priority date of the relevant patent application or patent.

Therefore, whether one is trying to determine whether an invention claimed in a patent is novel, inventive, or sufficiently disclosed, whether the claims are supported, or to determine the scope of a claim to ascertain whether a particular act infringes the claim, the PSA needs to be involved.

How is a PSA used?
So, how does one establish what a prior art document discloses to a PSA? Or what a term in a claim means to a PSA? Or what was the common general knowledge of a PSA working in a particular area before the priority date of a patent? One consults a PSA and obtains evidence from them.

Depending on the ground that is to be relied on and the nature of the evidence that therefore needs to be gathered from the PSA, the PSA may, for example, be asked a series of questions to ascertain their common general knowledge. They may be presented with relevant prior art documents and/or the patent or patent application. They may be asked what these documents disclose to them, and/or they may be asked to conduct prior art searching to show that particular documents would have been found by a PSA. They may even be asked to recreate experiments described in prior art documents, or in the patent or patent application itself.

The PSA then works with lawyers and/or patent attorneys to prepare the relevant declarations and/or affidavits in which they attest to certain facts and knowledge. The aim of the evidence is to assist the relevant audience (usually the Commissioner of Patents or a Judge) in determining exactly what a PSA would have done, thought or found given the circumstances and context.

Are we picking the right PSA?
This brings me to my main question: why do patent attorneys and lawyers select the experts that they do to represent the PSA? As mentioned above, according to well-established legal principles, the PSA should be a skilled but non-inventive worker in the relevant field of technology, who knows the common general knowledge in the art at the relevant time.

Given this, why is it that, for court proceedings in particular, the most highly qualified and therefore arguably least noninventive experts are selected? As eloquently enunciated by her Honour:5

The understandable tendency of a litigation lawyer is to retain the best, most renowned experts in the field. But the knowledge (and inventive capacity) of these experts might be far removed from the person skilled in the art. The greater the knowledge of the expert, the more they might have to disregard as not forming part of the common general knowledge at the priority date. Without detailed and explicit instructions, this is a strange position for an expert to be put in. They are being retained because they are renowned in the field, but the evidence being sought from them is not evidence of their own knowledge and capacities and what they could have invented – what is sought is their opinions about what effectively everyone in the relevant field would know and take as given.

Other artificialities arise as a result of selecting the most renowned expert in a field. Typically, such an expert has not relied on or used the common general knowledge that existed before the relevant priority date for many years, if not decades. Therefore, what they remember as forming part of the common general knowledge at a particular time may be difficult for them to accurately recall. In terms of re-creating and conducting experiments, many renowned experts have not actually been in a laboratory for decades. Whether they accurately remember how certain steps were performed before the priority date may therefore also be questionable.

It is interesting to note that, for the purposes of patent oppositions before the Australian Patent Office, the experts used often more accurately represent the PSA: they are often early-career scientists or engineers under the employ of one of the parties, who are still practically involved in conducting routine testing and experiments. In addition, these experts were often undergraduate students at the relevant priority date, and therefore their common general knowledge is representative of a PSA. Equally, during prosecution of a patent application, Australian patent examiners are often not experts in the field to which a particular technology pertains but are well-versed enough in that field to understand the technology and terminology, as well as have an appreciation of the relevant common general knowledge. Once again, the examiners also approximate a PSA more closely than renowned experts used in litigation proceedings. Therefore, there appear to be different PSAs used to assess the validity of a patent, depending on which stage of its life a patent is at.

The PSA in the era of Artificial Intelligence (“AI”)
Another conundrum about the identity of the PSA arises when we consider inventions made with the assistance of AI. If we take the early stages of drug development and identifying a candidate drug as an example, until quite recently the process has typically involved identifying a biological target (e.g., a receptor), followed by conducting in silico studies and/or screening compound libraries to find initial lead candidates that may act on that target. With the advent of AI, both steps (and particularly the second step) can be carried out much faster – once an AI system has been taught what sort of molecules may fit into and interact with a particular target, it can very quickly sort through vast numbers of compounds in a library to identify the optimum candidates. In this situation it is still scientists who have inputted the data to train the AI, and who have developed the library, and therefore the applicable common general knowledge is derivable in this context.

However, what happens in the case of more sophisticated AI systems, where a target is presented, and the AI is left to train itself and solve the problem of finding a drug candidate based on whatever data it can access? In this situation, it is no longer trained by humans that themselves are “constrained” by the common general knowledge in the field of their endeavour, and it can also access information from multiple (and sometimes not readily ascertainable) sources. The AI may produce results that are so far out of the common general knowledge, that an assessment of what a human PSA would have done in the context of trying to solve the problem becomes meaningless. This has implications not only for who the PSA is and what common general knowledge they possess, but also for legal tests, such as certain tests for inventive step, where, with no expectations or set course, it may no longer make sense to ask, in the words of the High Court of Australia:6

Would the person skilled in the art, when faced with the same problem, be directly led as a matter of course to try the claimed invention in the expectation that it might well produce a useful result?

Therefore, in addition to questions of inventorship, we will soon need to grapple with the implications of AI in other aspects of patent law. The PSA is already more than ordinary in certain contexts, and may soon transcend the human realm altogether!

This article was first published in Intellectual Property Forum.7


1 Justice Jayne Jagot, ‘Some Evidentiary Issues in Patent Law’ (2022) 130 Intellectual Property Forum 9.

2 Merriam-Webster (online at 11 January 2022) ‘reasonable person’.

3 Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183, 235; Stanway Oyster Cylinders Pty Ltd v Marks [1996] FCA 527; (1996) 66 FCR 577, 582.

4 Justice Jayne Jagot, ‘Some Evidentiary Issues in Patent Law’ (2022) 130 Intellectual Property Forum 9, 9.

5 Justice Jayne Jagot, ‘Some Evidentiary Issues in Patent Law’ (2022) 130 Intellectual Property Forum 9, 12.

6 Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59.

7 Dr Milena Dryza, ‘Not So Ordinary: Who is the “Person Skilled in the Art?”, 131 (March 2023) Intellectual Property Forum, 51.