Emotional Perception AI v Comptroller: The UK Supreme Court Reshapes UK Patent Law
How does the decision change how UK law assesses the patentability of computer implemented inventions and what will be…
Read more
The UKIPO has recently published its long-awaited guidance on the search and examination of patent applications following the Supreme Court’s decision in Emotional Perception AI Ltd v Comptroller-General of Patents, Designs and Trade Marks [2026].
The headline change is clear: Examiners should no longer apply the well-known Aerotel test when considering excluded subject matter. Instead, they must use the three-step approach endorsed by the Supreme Court. This brings UK practice closer to that of the European Patent Office (EPO), particularly for AI and other computer-implemented inventions. The same guidance would presumably apply to any mixed-type invention comprising both technical and non-technical features.
It does not, however, mean that adding a computer to a claim will make an otherwise non-technical idea patentable. The threshold question has become easier, but the real argument is likely to move to inventive step and, in particular, to which features of the claim contribute to its technical character.
The guidance follows the Supreme Court’s conclusion that the approach established by the Court of Appeal in Aerotel Ltd v Telco Holdings Ltd should no longer be followed, in light of the Enlarged Board of Appeal’s decision in G 1/19.
This is a significant departure from established UK practice. In the past, the UKIPO used Aerotel to decide whether an invention fell within one of the exclusions in section 1(2) of the Patents Act 1977, including the exclusion of programs for computers “as such”.
Examiners have now been told not to follow sections 1.07 to 1.51.7 of the current Manual of Patent Practice. Those sections are to be updated in due course. In the meantime, examiners must apply the Supreme Court judgment and the new guidance. Existing objections based on Aerotel should also be reconsidered afresh where an applicant responds to them.
The guidance describes a three-step assessment: a first hurdle, an intermediate step and a second hurdle.
The first hurdle asks whether the claimed subject matter is an “invention” at all. This is assessed using the “any hardware” test and without reference to the prior art. If the claim involves any technical means (for example, a computer, processor, server or computer-readable storage medium) it will pass this hurdle. The guidance says that an objection under section 1(2) should be raised only where the claim does not involve any technical means.
For most normally drafted computer-implemented method and system claims, this should be a low bar. A claim expressly requiring a computer will generally be enough. That is an important change from the previous UK approach, under which the question of excluded subject matter often depended on the substance of the alleged contribution.
Passing the first hurdle is only the beginning. The examiner must then carry out the new intermediate step and decide, feature by feature, which parts of the claim contribute to the technical character of the invention.
A feature contributes to the technical character where it contributes to, or interacts with, the technical subject matter of the claim in providing a technical solution to a technical problem. The feature is not considered in isolation. Its role must be assessed in the context of the invention as a whole. This matters because a feature that appears non-technical on its own may still make a technical contribution when combined with the other features of the claim.
The features that survive this filtering are then relevant to inventive step at the second hurdle. Examiners must continue to apply the established Pozzoli test. Where the only differences between the inventive concept and the prior art do not contribute to technical character, the claim will lack an inventive step. Where the differences do contribute, the examiner must go on to consider whether they would have been obvious to the skilled person.
In practical terms, the first hurdle deals with eligibility and should now be relatively easy to satisfy for computer-implemented inventions. The more difficult question is whether the features said to make the invention new and inventive also make a technical contribution.
The guidance gives less assistance on this central issue.
The Supreme Court said that the intermediate step must be carried out by reference to the technical character of the invention viewed as a whole. It suggested that identifying the inventive concept under Pozzoli might be an appropriate way of doing this in a particular case, but stopped short of prescribing that approach.
The UKIPO guidance similarly does not provide a UK-specific test or worked examples. Instead, it says that examiners may refer to Part G, Chapter II, section 3 of the EPO Guidelines when deciding whether a feature contributes to technical character. Those Guidelines deal, among other things, with the technical implementation of algorithms, the processing of physical data and computer programs that produce a further technical effect.
That reference will be useful, particularly given the limited UK case law on technical character. However, the EPO Guidelines are not binding on UK examiners. There may therefore be a period of uncertainty while UKIPO practice develops.
There is also an unresolved question about the relationship between the “technical character of the invention, viewed as a whole” and the “inventive concept” under Pozzoli. The two may be very similar in some cases, but the guidance does not say that they are necessarily the same.
The guidance does not clearly say whether the intermediate step should be carried out before or after the relevant prior art has been identified.
This is more than a procedural point. If technical character is assessed first, the examiner identifies the contributing features before carrying out the inventive-step analysis. If it is assessed in light of the prior art, the focus may instead be on whether the features that distinguish the claim from what is already known make a technical contribution.
Paragraph 39 of G 1/19 describes the intermediate step as filtering features that contribute to a technical solution to a technical problem in view of the closest prior art. The practical section of the UKIPO guidance, however, does not expressly require the prior art to be identified before the filtering exercise. It simply directs examiners to identify the contributing features between the first hurdle and the assessment of inventive step.
Further UKIPO decisions, and possibly further consideration by the courts, will be needed to settle the position.
Before Emotional Perception, UK examiners sometimes declined to carry out a substantive search where a computer-implemented invention was considered clearly excluded under Aerotel. This could be particularly unhelpful for UK start-ups using a first UK filing to obtain an early indication of the prior art before deciding whether to file abroad.
The removal of Aerotel, together with the low “any hardware” threshold, should reduce the number of applications refused a search on eligibility grounds. However, the new guidance does not expressly guarantee a full search merely because a claim refers to a computer.
Instead, it states that a search examiner should identify which features contribute to the technical character of the claimed subject matter and ensure that the search covers those features. This leaves open the possibility that, where an examiner regards most of the claim as non-technical, the search may be limited to a relatively small number of features.
We will therefore need to see how the guidance is applied in practice before knowing whether it results in more useful searches for computer-implemented inventions.
The guidance marks a substantial change in UK practice and should make the eligibility hurdle easier to overcome for AI and other computer-implemented inventions. It has not removed the underlying debate about what makes such an invention technical. That question has instead moved to the centre of the novelty and inventive-step analysis, where UKIPO practice is still taking shape.
This article was written by Senior Associate, Weronika Raszewska. If you would like to discuss this topic or anything else, please get in touch at weronika.raszewska@wilsongunn.com or contact one of our attorneys for advice.