Schrödinger’s Paradox in the Patent World: Can a Marketed Product Fall Outside the Prior Art?
On 2 July 2025, the Enlarged Board of Appeal of the European Patent Office (“EPO”), in its decision G 1/23, clarified whether a product already placed on the market, but whose internal structure cannot be analysed or reproduced by the skilled person, may nevertheless be excluded from the state of the art in the assessment of inventive step. In doing so, the Board added a new dimension to the ongoing debate following decision G 1/92.
The dispute giving rise to the decision concerned opposition proceedings initiated by Borealis against European patent EP 2 626 911 (“EP911”), owned by its competitor Mitsui Chemicals, which related to a solar cell encapsulant material and a solar cell module. Borealis argued that ENGAGE®️ 8400, a polymer marketed by the patentee, was commercially available at the filing date of EP911 and therefore had to be taken into account as prior art in the assessment of inventive step.
Although both parties agreed that ENGAGE®️ 8400 was a complex polymer, that its manufacturing process was not publicly available, and that the complete reproduction of a complex polymer (even if its final product could be analysed by the skilled person) was not straightforward, they diverged in their interpretation of the much-debated G 1/92 holding: “The chemical composition of a product is considered part of the state of the art once the product itself has been made available to the public and can be analysed and reproduced by the skilled person.”
The patentee, relying on a narrow reading of G 1/92, argued that since ENGAGE®️ 8400 could not be fully reproduced by the skilled person, it should not be considered part of the prior art at all. Borealis, by contrast, contended that G 1/92 should be interpreted more broadly: any characteristics of a product that was publicly available before the filing date and could be analysed, should form part of the prior art, even if the product could not be fully reproduced.
Although the Opposition Division initially rejected Borealis’s inventive-step objections, the case was first taken to the Board of Appeal, which decided to further refer it to the Enlarged Board of Appeal.
In its assessment, the Enlarged Board of Appeal rejected the idea that a product already on the market could be treated “as if it did not exist,” noting that such an approach would not reflect practical reality. The Board held that G 1/92 required broader interpretation. Whether a marketed product can be reproduced is, according to the Enlarged Board, a matter of proof to be assessed in each individual case; but the mere fact that it cannot be reproduced does not justify excluding it from the state of the art.
Most importantly, with G 1/23, the test of “analysability + reproducibility” was abandoned. The Enlarged Board considered that analysability (i.e., the ability of the skilled person to determine the product’s composition/properties through reasonable effort) is sufficient for the product to belong to the state of the art. Reproducibility is no longer a mandatory requirement. According to the decision, if a product is publicly available and the skilled person can understand its features—wholly or in part—that knowledge forms part of the prior art. Moreover, “reproduction” must be understood broadly: it does not mean that the skilled person must be able to manufacture the product themselves, but rather that they must be able to obtain or possess it (for example, by purchasing it). Thus, a marketed product cannot be excluded from the prior art merely because its full composition or structure cannot be completely analysed or identically reproduced.
In September, The Board of Appeal applied G1/23 for the first time in its decision numbered T 0143/24. The decision referred for appeal was rendered by the Opposition Division before G 1/23 came out and the Division had reasoned that, even though the applicant presented two polyethylene resin products that were already on the market before the priority date, these products did not belong to the state of the art, as there were no public disclosure explaining the products or their means of manufacture.
During the appeal review, the Board of Appeal implemented the G 1/23 principle to the case at hand and reached the conclusion that the products presented by the applicant should be considered as the closest prior art, regardless of the absence of detailed explanations on how to manufacture them and considered in the novelty and inventive step examinations.
The finding of G 1/23—that products already on the market but not fully reproducible by the skilled person are nevertheless prior art—has undeniable potential to reshape both claim-drafting practices and even marketing strategies of patent proprietors. Furthermore, it is indisputable that companies investing in R&D will need to re-examine the fine line between keeping product development as a trade secret and filing a patent application. It is beyond doubt that this decision will reverberate in the patent world for years to come.
First published by Legalink Newsletter in Sep 30, 2025.