- Insights
- The IP Brief - October 2026
The IP Brief - October 2026
Download PDF VersionWelcome to our latest edition of The IP Brief – a quarterly update of key IP cases and news, with a primarily UK and EU focus.In this edition, we take a look at:
- the UK Supreme Court’s landmark decision on jurisdiction to determine FRAND rates for patent pool SEPs;
- the CJEU’s finding, in Anne Frank Fonds (Case C-788/24), that there will be no communication to the public, for copyright purposes, where effective website geo-blocking is used to restrict access to copyright works in protected countries;
- exhaustion of copyright in “complex” computer software, following the UK Court of Appeal’s decision in JJH Enterprises v Microsoft;
- the ongoing copyright infringement dispute between Shein and Temu relating to photographs for product listings;
- whether freedom of expression may constitute “due cause” for using a third party’s trade mark without consent;
- the UK High Court’s approach to assessing damages for trade mark infringement in Montres Breguet v Samsung; and
- the meaning of “public domain” in confidentiality agreements, following the UK Court of Appeal’s decision in IlliquidX v Altana Wealth.
Patents
Cleaning the (patent) pool: Supreme Court opens the door to judicial determination of FRAND rates for patent pool SEPs
In another example of the English courts’ willingness to accept jurisdiction to determine fair, reasonable and non-discriminatory (FRAND) rates for standard essential patents (SEPs), the Supreme Court has recently handed down its decision in the long-running case of Tesla v InterDigital, following Tesla’s appeal. In it, the court unanimously confirmed the ability of the English courts to determine whether the terms of a licence offered by patent pools and platforms (Patent Pools) that include SEPs are FRAND. This is the case even where, as here, the operator of the Patent Pool has made no FRAND commitment and is not the owner of any of those SEPs. The decision is an important one as it paves the way for judicial scrutiny of the rates being charged by operators for their Patent Pools.
Background
The Patent Pool in question covers various 2G-5G SEPs, for which Tesla needed a licence in order to implement 5G connectivity in their vehicles. It was operated by Avanci, as agent on behalf of (at the time) approximately 65 SEP owners, who had authority to license the SEPs in a pre-determined global manner at a rate of $32 per vehicle. Avanci considered this rate to be FRAND, but Tesla thought it too high. Tesla therefore sought declarations from the court against both Avanci and InterDigital (as the owner of certain key UK SEPs on the platform) that Tesla is contractually entitled to a FRAND licence to all SEPs on the platform and that the rate offered by Avanci was not FRAND and should be set at a lower rate determined by the court (the “FRAND claims”).
The High Court initially struck out the FRAND claims on jurisdiction grounds, largely on the basis that it found that there wasn’t a serious issue to be tried. It also, obiter, agreed with the respondents that Tesla had failed to show that the English courts were clearly a more appropriate forum to hear the claim than the Delaware Court of Chancery (InterDigital and Avanci both being Delaware entities).
The majority of the Court of Appeal dismissed Tesla’s appeal, agreeing with the High Court. However, Arnold LJ dissented, stating he would have allowed the appeal both in relation to the question of whether there was a serious issue to be tried and whether the Delaware Court of Chancery was an available forum for the FRAND claims.
Supreme Court’s decision
The Supreme Court essentially agreed with Arnold LJ. It concluded that there is a serious issue to be tried as to whether the FRAND obligation applies to an offer to license jointly through a patent pool or platform (as opposed to bilaterally). The court was clearly concerned that if the FRAND obligation fell away when SEPs were licensed through a Patent Pool, it would create an obvious loophole in the FRAND system. The court also found that there was nothing in the FRAND obligation or the broader IPR policy underpinning the ETSI standards regime that indicated that the FRAND obligation shouldn’t apply in that scenario.
Assuming that the FRAND obligation does apply, the Supreme Court went on to find that there is also a serious issue to be tried as to whether a FRAND licence of InterDigital’s UK SEPs would have to be a global licence to the whole Avanci 5G platform at a FRAND rate. In reaching this conclusion, the court noted that if the commercial reality of licensing standards such as the 5G standard is that bilateral licensing is not practicable due to the sheer number of SEP owners involved “that is a powerful factor in support of the conclusion that only a platform licence can be FRAND”. Indeed, the practice of Patent Pools has developed to deal with this issue and the court noted that some licensors relied on the availability of the Patent Pool licence to discharge their FRAND obligation.
Finally, the Supreme Court considered that there was a reasonable prospect of the English courts granting the declarations sought by Tesla and that there was no need to join the other SEP owners.
As for the other jurisdictional challenges, the Supreme Court agreed with Arnold LJ that the claim should be characterised as a claim about the licensing of InterDigital’s UK SEPs on the Avanci 5G platform. On that basis, the court agreed with Arnold LJ that the Delaware Court of Chancery would likely not determine a FRAND rate based on non-US patents (such as those UK SEPs) and therefore the English courts were the more appropriate forum.
What this means for SEP owners and implementers
Importantly, the Supreme Court’s decision does not conclusively determine the issues highlighted above in substance as they have been referred back to the High Court for full trial. However, the decision does clarify that the English courts have jurisdiction to consider Patent Pool rates involving UK SEPs and opens the door to them determining what the FRAND terms would be for collective licences. That is good news for implementers as it reflects the commercial reality of licensing standardised technology (which is becoming increasingly critical in a world of interconnected devices) without losing the right to benefit from FRAND terms. It does, though, raise questions for SEP owners who license their patents via a patent pool or platform, as well as for pool and platform operators, who will all need to consider whether or not the rates they are setting are truly FRAND.
Regardless of what is ultimately decided, it seems likely that there will be an uptick in claims from implementers asking the English courts to consider whether the rates set by Patent Pools are FRAND and also that SEP owners will want greater control and involvement in the rate-setting and licensing practices of their Patent Pool operators going forward. That, in turn, may well have a wider impact on the operation of Patent Pools and the broad range of businesses that rely on them to access key technology.
Copyright
CJEU confirms no communication to the public where effective geo-blocking used
The CJEU has confirmed, in Anne Frank Fonds (Case C-788/24), that there will be no communication to the public, for copyright purposes, where effective website geo-blocking is used to restrict access to copyright works in those countries where the works are protected. This is so even if users can circumvent the geo-blocking measures using a virtual private network (VPN) or similar service.
Background
Anne Frank Fonds (the Fund) owns the copyright in certain parts of Anne Frank’s Diary in the Netherlands. That copyright lasts until 2037. The Diary is, however, already in the public domain in several other EU member states, including Belgium (PD Countries), with copyright having expired in those countries in 2016.
In 2021, the defendants published a new, scholarly edition of the Diary online, with access geo-blocked to restrict availability to the PD Countries. Any user who tried to access the Diary in the Netherlands was presented with an access denied message, and users accessing the website from elsewhere were required to confirm they were in a PD Country.
In spite of this, the Fund brought a copyright infringement action against the defendants before the Dutch courts, arguing that individuals could still access the Diary in the Netherlands using a VPN. The claim was initially dismissed by the Dutch courts, but the case progressed all the way up to the Netherlands Supreme Court which referred three questions to the CJEU. These can be summarised as follows:
- Does online publication of a work have to be targeted at a particular country in order for there to be a communication to the public in that country under Article 3(1) of the InfoSoc Directive (Directive 2001/29/EC)? If so, what factors must be considered in that assessment?
- Can there be a communication to the public in a particular country if, by means of (state-of-the-art) geo-blocking, it is ensured that the website on which the work is published can only be accessed by the public in that country by circumventing the blocking measure using a VPN or similar service?
- If question 2 is answered in the affirmative, is that communication made by the party who published the work online?
AG Rantos’ Opinion
As we reported back in February (see our blog), AG Rantos’ opinion was that an online publication does not need to be addressed to the public in a specific country in order to constitute a communication to the public in that country. All that is required is that there is an “act of communication” of a work and that that act of communication is to a public.
He then went on to find that where content is published on a website which is subject to effective, state-of-the-art geo-blocking measures, there will be no communication to the public in those blocked territories.
That being the case, he did not need to answer the third question. However, he noted that if the possibility of circumventing geo-blocking measures did amount to a communication to the public, a VPN service provider could not be held liable for the acts of those who use its services to circumvent geo-blocking restrictions. In that scenario, the VPN service provider would simply be acting as an intermediary, unless they actively encourage such unlawful use.
CJEU decision
While the thrust of the CJEU’s decision aligns with AG Rantos’ Opinion, the court reformulated the first two questions in such a way that it didn’t, in fact, expressly address the first question about whether targeting a particular country is required in order for there to be a communication to the public in that country.
Instead, it focussed on answering the combined question, “whether Article 3(1) of [the InfoSoc Directive] must be interpreted as meaning that a work which is in the public domain in certain Member States, but is still protected by copyright in another Member State, and which is published free of charge on a website that incorporates a geo-blocking measure designed to ensure that access to that site is blocked for internet users visiting it from that other Member State, is the subject of a ‘communication to the public’ in that Member State, within the meaning of that provision, when those internet users can circumvent that geo-blocking measure using a VPN or similar service.”
The court’s answer to that question was no, provided that the geo-blocking measure in question is “effective”. In the context of this case, that meant that the geo-blocking measure must be “state-of-the-art” and suitable for achieving the objective of preventing access in those countries where the Diary is still protected by copyright. The fact that a geo-blocking measure can be circumvented (for example, via a VPN) is not decisive in this respect, nor is the fact that other technological measures may take greater account of the copyright holder’s interest (there being a need to balance the copyright holder’s interests with those of the public in enjoying access to the works in PD Countries).
The court made clear, however, that simply asking the user to confirm that they’re accessing the site from a country in which the work is not protected by copyright will not be sufficient.
Turning to the third question, and given the court’s finding that whether there is a communication to the public in this context depends on the effectiveness of the geo-blocking measures that have been put in place, the court answered this from the perspective of cases where the measures used are ineffective rather than where the measures used can be circumvented. In those circumstances, the court found that the communication would be made by the person who published the work.
Comment
This decision will be welcomed by those looking to publish content online in a copyright-compliant manner. Had the court found otherwise, it would have made it very difficult to share works online without infringing copyright, where copyright has expired in some countries but not others, given no geo-blocking measures will be 100% effective.
It’s important to note, however, that the judgment goes beyond finding that there will be no communication to the public where effective website geo-blocking is used. Indeed, the CJEU has seemingly imposed an obligation on online publishers who know (or ought reasonably to know) that a work is protected in some EU member states but not others “to adopt effective technological measures to restrict access to the website solely to those internet users [based in PD Countries], failing which, given the freely accessible nature of websites, that person would be giving access to the work concerned to all internet users and would thus be infringing that right”.
Perhaps unfortunately, due to the way the court reframed the questions before it, the CJEU didn’t directly answer the first question referred (about whether targeting a particular country is required in order for there to be a communication to the public in that country). AG Rantos had earlier opined that such targeting is not required (see our blog), but the CJEU did not explicitly confirm or deny that view. One could, however, read the decision (including the sentence quoted above) as implicitly agreeing with AG Rantos that targeting is not required in order for there to be a communication to the public in the EU. If that approach is adopted going forwards, it could lead to divergence with the UK on this point, with the English courts having noted on a number of occasions (including in the High Court’s recent decision in the Peppa Pig and Wolfoo dispute that was handed down on 25 June) that targeting is required in the UK.
JJH Enterprises v Microsoft: exhausting another avenue in the battle over software resale markets
As the software resale market continues to develop, questions remain over when holders of UK and EU copyrights might be able to prevent sales of second-hand software licences.
More than a decade after the CJEU’s landmark 2012 ruling in UsedSoft v Oracle, the UK Court of Appeal has recently delivered a comprehensive judgment on the application of the principle of copyright exhaustion to complex computer software, addressing some of these questions in the process.
The facts
JJH Enterprises (trading as ValueLicensing) (VL) is in the business of selling second-hand licences for Microsoft software products such as Microsoft Office and Microsoft Windows.
VL claimed that Microsoft had stifled the resale of those licences through its actions in migrating customers from a perpetual licensing model to a subscription-based one, and by including certain restrictions in the underlying enterprise agreements, between 2014 and 2022. It brought a first-instance claim in the Competition Appeal Tribunal (the CAT) against Microsoft, arguing such actions breached competition law in the UK and the EEA (including the EU).
Part of Microsoft’s defence hinged on copyright law. It sought to argue that the copyright in its software products had not been exhausted, that VL had no right to resell the relevant licences in the first place and that VL was in fact itself infringing Microsoft’s copyright. If Microsoft were right on these points (which the parties agreed were to be determined under EU law in force at the time, in the form of the Software Directive, the InfoSoc Directive and relevant CJEU case law), it was common ground that VL’s competition claim would fail.
Microsoft also challenged the CAT’s jurisdiction to adjudicate the copyright issues, arguing that those issues should be determined by the High Court. The CAT found that it did have jurisdiction to hear the copyright aspects of the claim where (as here) doing so was necessary to resolve the competition claim before it, with the two substantive copyright points later being tried as preliminary issues.
Issues at stake
Two core copyright questions were at the heart of this case.
The first concerned the correct approach to digital exhaustion for so-called “complex matter” – that is, where works comprise both a computer program protected by the Software Directive and other non-program copyright works (such as icons and user interfaces) that are protected under the InfoSoc Directive.
The second looked at whether VL could legitimately sub-divide bulk, or volume, licences sold by Microsoft and then resell them. For example, if Microsoft sells 1,000 licences in bulk, can those licences be split up so that VL can sell on a portion (e.g., 500) whilst keeping the rest (e.g., the remaining 500)?
At first instance, the CAT held that all copyright, including that of the non-program works, in the resold software products had been exhausted by virtue of their first sale by Microsoft. It also found that VL’s sub-division of the relevant licences was permissible.
Microsoft appealed both points (as well as the CAT’s jurisdiction to determine the copyright aspects of the claim) and the Court of Appeal handed down its judgment on 7 July.
The Court of Appeal’s decision
The Court of Appeal upheld the CAT’s finding that it had jurisdiction to hear the copyright aspects of the claim.
Perhaps more importantly, however, is the Court of Appeal’s judgment on the two core copyright questions noted above.
Approach to digital exhaustion for “complex matter”
As some readers will be aware, exhaustion of rights is treated differently under the Software Directive and the InfoSoc Directive. Under the Software Directive, first sale of a copy of a computer program by the rightsholder (or with their consent) exhausts the distribution right in that copy. This applies equally to tangible and intangible (i.e., downloaded) copies, as per the CJEU’s decision in UsedSoft v Oracle. The distribution right under the InfoSoc Directive, by contrast, is only exhausted on first sale of a tangible object. So, as an example, whilst the distribution right would be exhausted on first sale of a physical book, it would not be exhausted on the first sale of an e-book (as per the CJEU’s decision in Tom Kabinet).
Going back to the Microsoft case, the products in question were made up of computer programs (governed by the Software Directive) and certain non-program works, such as graphic works in the form of icons and user interfaces (governed by the InfoSoc Directive).
The Court concluded that the correct approach to dealing with these types of “complex” products is to consider the product as a whole and to make a judgment about its proper characterisation in order to determine which single regime governs it (Software or InfoSoc Directive), rather than apply both regimes to their respective components.
Applying this to the Microsoft Office and Windows products in issue, the Court of Appeal agreed with the CAT that these are, in substance, computer programs, with the graphics, user interface and similar works being incidental or having only accessory character. As a result, the Software Directive was found to apply and Microsoft’s copyrights in both the program and non-program works were exhausted on first sale.
Division of bulk licences
The Court of Appeal also upheld the CAT’s decision on the second copyright issue, finding that Microsoft’s bulk licences could lawfully be sub-divided.
Microsoft sought to argue that the CJEU had already found in UsedSoft that this sort of sub-division is not permitted. However, the Court of Appeal agreed with the CAT that UsedSoft could be distinguished from the facts before it. In UsedSoft, a single copy of the software was downloaded and stored on a central server, with that copy then accessed by individual users in a “hub and spokes” arrangement. Here, however, each copy of the software was found to be independent of the others, with no program running on a central server.
The Court of Appeal also found that it makes no difference whether the software being licensed is contractually labelled as a single licence for multiple devices or as a bundle of separate licences. What really matters is what the first acquirer does with the original copies after resale. If they keep and retain one or more of the original copies (such that there are more copies in circulation than originally licensed), those retained copies will be unlicensed and infringing. As the Court of Appeal made clear, however, any such failure by the first acquirer will not affect a later purchaser’s right to use the copy it has bought.
Practical takeaways
The Court of Appeal’s decision is a helpful clarification of how copyright exhaustion principles apply in a digital software context.
As the Court explained, complex products need to be considered in the round to determine which exhaustion rules (Software Directive or InfoSoc Directive) apply. Software providers should take note that bundling non-program works – such as icons, user interfaces and fonts – with program works will not, by itself, prevent exhaustion.
When it comes to bulk software licensing, the Court of Appeal has made clear that there is no blanket ban on sub-division. Software vendors seeking to control the resale of their products should therefore carefully look at how their software deployment is structured. Contractual labels alone will not determine the outcome.
With public market research projecting strong growth in software resale markets over the years to come, there is a lot at stake here and it is clear to see why this is an important battleground for rightsholders and second-hand licence vendors alike. This remains a case to watch, with Microsoft reportedly looking to appeal to the Supreme Court.
SHEIN vs Temu: Marketplace copyright infringement claim comes apart at the seams
In August, the UK High Court dismissed the entirety of a copyright claim brought by the e-commerce platform Shein against a rival e-commerce platform, Temu, over the use of certain photographs of Shein-branded clothing on Temu’s UK website.
The judgment covered extensive ground, but the court’s (obiter) finding that Temu could have relied on the hosting defence was particularly striking at a time when that defence has faced pressure elsewhere. The case also offers a cautionary tale about the importance of establishing chain of title before commencing copyright infringement proceedings.
Background and chain of title issues
Shein alleged that Temu had infringed its copyright in photographs used in 2,559 product listings that had been uploaded to Temu’s UK site by third-party merchants (many of whom were also Shein suppliers). The listing photographs fell into three buckets: those taken by Shein employees; those supplied to Shein by agencies (or freelance photographers); and those provided by Shein’s suppliers.
Temu denied infringement and, in the alternative, argued that if infringement was found, it could avail itself of the hosting defence in Regulation 19 of the E-Commerce Regulations.
To manage the scale of alleged infringements, the legal issues were to be tried via a sample-set of 20 product listings (the “trial sample”). By the time of trial, however, Shein had abandoned its claims in all but five of this set, largely due to difficulties in establishing a chain of title to the works, leaving four employee works and a single supplier work referred to as the “Strawberry Nightdress”.
Even then, Shein was found to have initially only held a non-exclusive licence to use the “Strawberry Nightdress” photographs. Shein and the relevant rightsholders had then executed a further agreement as a “catch-up” assignment, pursuant to which Shein also acquired equitable title to accrued rights of action. But, in the end, that still wasn’t enough to save Shein’s claim for copyright infringement in relation to the Strawberry Nightdress photos, as the Temu seller in question was found to have been given consent to use them.
No infringement of employee works
The Court also rejected each of Shein’s infringement claims in relation to the employee works. The key findings were as follows:
- Copying and authorisation: It was common ground that Shein could not bring a successful copyright infringement claim against Temu based on copying as Temu’s servers are located outside the UK. Any copying of the relevant photographs by Temu would therefore have been outside the territorial scope of UK copyright law. Instead, Shein argued that Temu had authorised users of its UK website to reproduce the trial sample photographs by viewing them on their browsers. That argument was given short shrift by the court, which held that such copies fell within the temporary copies exception in s.28A CDPA. Crucially, the court considered the temporary copies to have no independent economic significance to the website user, with the judgment describing the photographs as “the online equivalent of looking at products in a bricks and mortar shop window”.
- Communication to the public: Shein also argued that Temu had communicated the trial sample photographs to a ‘new public’ by making them available on its UK website. That claim also failed as the Court found that Temu’s conduct did not amount to “a deliberate intervention, in full knowledge of the relevant consequences, to give the public access to protected works”. Interestingly, in reaching this conclusion, the High Court followed the CJEU’s decision in Peterson v Google (C-682/18), which was handed down post-Brexit, noting the relevance of that decision to the case before it and the desirability of retaining consistency with the CJEU’s approach.
- Secondary infringement: Finally, Shein argued that Temu had committed secondary infringement of its copyright by exhibiting and distributing the trial sample photographs to users of its UK website whilst knowing (or having reason to believe) that those photos were infringing copies. This claim also failed as the Court found that Temu did not know, and had no reason to believe, that the relevant photographs were infringing copies before they were removed.
Hosting defence
Having found no copyright infringement, the Court was not obliged to consider the hosting defence. The Court nonetheless took the opportunity to note that Temu could have successfully relied on it.
The hosting defence broadly protects platforms from financial liability for storing copyright-infringing material, provided they meet certain criteria, including playing a passive intermediary role. Having surveyed the key CJEU and English authorities, the Court concluded that the core question before it was not the degree of Temu’s control over its general platform and dealings in the abstract, but rather its precise role over the specific disputed content.
Shein pointed to Temu’s alleged involvement in product pricing, product quality and selection, promotions, logistics and customer service, but relied on evidence that the judge deemed “very weak” (with heavy reliance placed on hearsay, third party articles and blog posts). Furthermore, the judge noted that such arguments, even if accepted, missed the mark. They did not demonstrate Temu’s active role over the specific trial sample works, which was instead found to be “of a mere technical, automatic and passive nature”. And, in any event, once aware of any alleged infringements, Temu had acted expeditiously to remove the relevant product listings.
Comment
This is a fascinating case that provides insight into a number of topical issues.
First and foremost, it highlights the importance of establishing chain of title before commencing a copyright infringement claim. Reliance on contractual warranties given by third parties will not be sufficient.
The decision also highlights the risks of diving in too quickly, with Shein itself being found liable under its cross-undertaking in damages for product listings that were wrongly taken down following notifications from Shein under interim injunction orders that had been granted by the Court.
Perhaps most interesting, though, are the Court’s findings on the hosting defence. While the High Court’s judgment can provide some comfort to platform operators that the hosting defence still carries teeth in the UK, in the period between oral argument and judgment, the CJEU issued its own decision in WebGroup Czech Republic and Coyote System (C-188/24 and C-190/24), which has caused concern that the hosting defence in the EU may have somewhat come to an end for most large online platforms. This is because the CJEU’s headline finding in that case suggests that platforms that use algorithms to determine, in their own interest, how, when and in what order users’ content is broadcast or displayed, exercise sufficient control to fall outside the hosting exemption. The CJEU judgment is a notable contrast to the Temu case, in which the hosting defence-related arguments and analysis did not focus on such algorithms or so-called ‘recommender systems’.
Shein has, however, recently received permission to appeal. Platform providers will therefore be watching closely to see how these issues are addressed at the appellate level.
Trade Marks
Freedom of expression may be "due cause" for using a third party's trade mark
The Grand Chamber of the CJEU has ruled, in IKEA (C-298/23), that freedom of expression, including freedom to express political opinions and political parody, may constitute “due cause” for using a third party’s trade mark without consent.
The judgment provides helpful guidance on how EU national courts should balance the fundamental rights to freedom of expression and IP in a trade mark infringement context and the criteria to be taken into account when assessing that balance.
Background
IKEA, the well-known furniture and home accessories retailer, owns a number of registered Benelux and EU trade marks relating to the IKEA name.
A Belgian political party, Vlaams Belang, used IKEA’s name, typography and colour palette as part of its political plan for reforming the Belgian asylum and immigration policy, which it called the “IKEA-PLAN – Immigratie Kan Echt Anders” (Immigration Really Can Be Different). It also used illustrations containing the IKEA trade marks and characters resembling those used by IKEA in its product assembly instructions.
IKEA brought proceedings for trade mark infringement under Articles 10(2)(c) and 10(6) of the Trade Marks Directive and Article 9(2)(c) of the EU Trade Mark Regulation. Under each of these provisions, to establish infringement, it must be shown (amongst other things) that the defendant’s “use of [the] sign without due cause takes unfair advantage of, or is detrimental to, the distinctive character or the repute of the trade mark” (emphasis added).
The defendant acknowledged that it had used IKEA’s marks without consent, but it claimed that it had due cause to do so, in order to reinforce its message and increase its dissemination.
This ultimately led the Belgian court to refer a number of questions to the CJEU, asking in essence:
- whether freedom of expression, including the freedom to express political opinions and political parody, may constitute due cause for using a sign identical with or similar to a trade mark with a reputation within the meaning of Articles 10(2)(c) and 10(6) of the Trade Marks Directive and Article 9(2)(c) of the EU Trade Mark Regulation; and
- if so, what criteria should EU national courts take into account in assessing the balance between the right to freedom of expression and the right to IP?
CJEU decision
The CJEU found that freedom of expression may constitute due cause.
In reaching that conclusion, the CJEU noted that the phrase “due cause” is not defined in either the Trade Marks Directive or the EU Trade Mark Regulation. However, the recitals to both pieces of legislation expressly state that they should be applied in a way that ensures full respect for fundamental rights and freedoms, particularly freedom of expression. In those circumstances, the court found that trade mark owners’ rights may be limited by the need to protect a third party’s freedom of expression and, consequently, that freedom of expression may constitute due cause.
That left the court to determine what criteria the EU national courts should take into account when balancing the rights of freedom of expression and IP and, ultimately, in determining whether the case for due cause has been made out. It concluded that national courts must take into account all relevant circumstances of the case, including:
- the intention of the third party using the sign;
- whether the sign was being used in good faith (which will be the case where, for example, the sign is used to convey an idea or opinion which relates to the trade mark itself, the trade mark owner, their goods and services or commercial practices; or where the use is necessary for other reasons such as where the mark has become a public cultural reference or part of everyday language);
- whether the use contributes to a public interest debate and whether that use occurs in a strictly commercial context; and
- the consequences for the trade mark owner, taking into account things like the intensity, extent and methods of use of the sign, the scale of the mark’s reputation, the degree of similarity between the sign and the mark, and whether the use may give the impression that the trade mark owner agrees with the political message or ideas being promoted, or supports them in some way.
Ultimately, it will be for the Belgian national court to determine the outcome in this case. However, the CJEU judgment provides a fairly strong indication that, in its view, applying the above factors to the case before it, the CJEU did not believe there was due cause for the use made of IKEA’s marks. This was for a number of reasons, including that: the word IKEA has no semantic meaning that would justify the need for it to be used by Vlaams Belang; IKEA does not appear to have become a public cultural reference or part of the common language; the marks did not appear to have been used to engage in debate concerning the owner of the marks or its goods or services; the signs used were highly similar to or identical with the IKEA trade marks and were used repeatedly, including online; the use could create the impression that IKEA agreed with or supported the political message being made (which runs counter to IKEA’s aim to be politically neutral); and that the use was for the sole purpose of taking advantage of IKEA’s reputation.
Comment
This is an interesting decision which confirms that freedom of expression may provide due cause for using a third party’s trade mark without consent. Whether or not it does so will depend on the specific facts of the case in issue and so it will be interesting to see how EU national courts apply the judgment, and weigh the balance between freedom of expression and the right to IP, in future.
That said, the court was clearly keen to put some guardrails around this, noting that “mere reliance” by a party on their right to freedom of expression will not be sufficient to establish due cause. Instead, a party wishing to make this type of argument will need to set out the specific grounds for its use of the sign in question in connection with the exercise of its freedom of expression and demonstrate why those grounds should take precedence over the rights and interests of the trade mark owner.
Samsung ordered to pay Swatch Group $11.6mn in damages for trade mark infringement
The UK High Court has ordered Samsung to pay the Swatch Group $11.6mn following the Court of Appeal’s finding back in December 2023 that Samsung was liable for trade mark infringement where consumers had downloaded third party digital watch face apps from the Samsung Galaxy Appstore for use on Samsung’s smartwatches.
Background
Samsung makes a range of smartwatches, specifically marketed for their watch-like qualities, with the ability to download from the Samsung Galaxy Appstore (SGA), and display, third party branded digital watch faces (e.g. Tissot, Longines, Omega) for use on Samsung’s smartwatches. Some of the watch faces made available for download from the SGA were developed by third party app developers and made infringing use of trade marks owned by the Swatch Group. Keen to stop this, Swatch brought trade mark infringement proceedings directly against Samsung, arguing that although Samsung was not the developer of the infringing watch faces, it was intimately involved in, and controlled, the process by which the apps were made available to consumers.
At first instance, the High Court agreed with Swatch that most of the apps in question did indeed infringe the Swatch Group’s trade marks and that Samsung was primarily liable for those infringements. This was later upheld by the Court of Appeal (see our blog), with the infringements falling into three categories:
- use on the watch face (Dial Branding Infringements);
- use in an app name on the SGA (Store Display Infringements); and
- use in both an app name and on the watch face.
Infringement having been found, the High Court was left to determine the damages payable by Samsung to the Swatch Group.
The Swatch Group sought $170mn, based on its valuation of a complete co-branding licence under which the key elements of the Swatch Group’s brands would have been superimposed on Samsung smartwatches and related hardware and packaging. Samsung, on the other hand, argued that damages should be limited to $301, based on its valuation of a licence for the actual use made of the Swatch Group marks.
Approach to quantifying damages
The court agreed with the Swatch Group that damages were to be assessed on a “negotiation damages” basis (aka “licence fee”, “Wrotham Park” or “user” damages), but it was not persuaded that either party’s assessment of those damages was accurate.
Whilst calculating negotiation damages is relatively straightforward where there is a market for the claimant’s IP (and therefore comparables to reference), it is more difficult where there is no market, as was the case here – the Swatch Group did not license out its brands at all (it was “an unwilling seller”) and Samsung’s infringements had not diverted sales away from the Swatch Group.
In cases such as these, the court found that the objective is to determine the “economic value of the right in question”. The problem is how to determine what is in fact “an objectively reasonable assessment of value”.
Having reviewed the case law, the court pulled out a number of points of guidance. The starting point, the court said, is to determine what rights the infringer needs from the claimant in order not to infringe, with the subject matter of the hypothetical licence being “the specific infringing conduct that the infringer intends to carry out at the relevant time”. The licence scope should therefore be assessed on the basis of a reasonable person’s expectations as to the future, rather than with the benefit of hindsight. That will then inform what the infringer might reasonably be prepared to pay for those rights.
Damages payable by Samsung
Applying this to the facts before it, the first step for the court was to determine the terms of the hypothetical licence that Samsung would have needed, before the event, to make what it was proposing to do lawful.
The scope of the licence proposed by Samsung (based on actual use made of the Swatch Group marks) was found to be too narrow and based on hindsight, only partially covering the infringements that had been found. In contrast, the scope of the licence proposed by the Swatch Group (a complete co-branding licence) was found to be too broad and far more extensive than what Samsung actually needed in order to cover its infringements or would have contemplated at the time of the hypothetical negotiation. The judge therefore rejected the approach of both experts. Instead, the court found that what Samsung would have wanted was a licence to validate the infringements it could anticipate, i.e. to enable Samsung to operate the SGA without infringing the Swatch Group’s marks and to sell watches without causing users (and Samsung) to infringe when relevant watch face apps were downloaded; and on the basis that any use of the Swatch Group’s marks would be rectified by removal as soon as identified.
As for the hypothetical licence fee payable under that licence, the court found that there would have been separate licence fees for the Store Display Infringements and Dial Branding Infringements. It regarded the Store Display Infringements as the more serious infringements and, adopting a broad-brush approach, awarded a flat fee of $10mn for those. For the Dial Branding Infringements, the court ultimately landed on a flat fee of $10 per download, resulting in damages of $1.6mn. That resulted in a total award of negotiation damages of $11.6mn.
Comment
This case provides an interesting example of how the English courts will assess negotiation damages in circumstances where the infringer’s actions did not divert sales away from the trade mark owner and where the trade mark owner is not in the habit of licensing out its rights.
Given the huge difference in licence scope and valuation between the experts, and the lack of comparables to fall back on, the court was left in a difficult position. Those seeking negotiation damages of this nature in future will want to think carefully about the scope of the licence required and their valuation methodology in order to improve the chances of the court siding with their assessment.
Confidential Information
UK Court of Appeal opines on meaning of "public domain" in NDA
The UK Court of Appeal has confirmed, in IlliquidX v Altana Wealth, that, absent drafting to the contrary, references to “public domain” in confidentiality agreements will generally be interpreted in line with the expression’s established meaning in the law of confidentiality.
Background
IlliquidX, Altana and Brevent established a joint venture to set up a fund to invest in Venezuelan distressed debt. The parties entered into a JV agreement, together with a non-disclosure and non-circumvention agreement (NDA) based on a professionally drafted template.
The JV came to an end without launching a fund. Altana later set up its own fund focussed on distressed Venezuelan debt. IlliquidX claimed that, in setting up the fund, Altana and Brevent had misused its confidential information and trade secrets.
The NDA provided protection for broad categories of confidential information, subject to certain exceptions. In particular, the confidentiality undertakings did not apply to information which “at the time of supply is in the public domain” or “subsequently comes into the public domain other than as a result of breach of the undertakings contained in” the NDA.
Altana and Brevent accepted at trial that much of the information IlliquidX was seeking to rely on fell within the definition of “Confidential Information” in the NDA. However, they claimed that the public domain exception applied, arguing that “public domain” extended to any information which was available or had been disclosed without being subject to a duty of confidentiality, such that the exception would apply where (just) one other person was free in law and equity to use that information.
At first instance, the High Court found that the opportunity to set up a sanctions-compliant fund to exploit undervalued Venezuelan debt was confidential information within the meaning of the NDA, that it was not covered by the public domain exception and that Altana and Brevent had misused it.
Altana and Brevent appealed on a number of grounds, including: (i) the meaning of “public domain” in the NDA; and (ii) the judge’s conclusion that none of the information relied on was in the public domain.
Meaning of “public domain”
The Court of Appeal agreed with the High Court that where parties to a professionally-drafted contract use an expression with an accepted meaning in law, that expression will be presumed to bear that meaning unless the wording of the contract indicates that the parties intended otherwise.
The phrase “public domain” has a well-established meaning in the law of confidence - information is in the public domain where it is “so generally accessible that, in all the circumstances, it cannot be regarded as confidential”.
The Court of Appeal went on to find that the references to “public domain” in the NDA had to be construed in accordance with this meaning. There was nothing in the wording of the NDA or the wider factual matrix to suggest that the parties intended “public domain” to bear any other meaning; nor was the court persuaded by the appellants’ argument that the public domain exception should be interpreted broadly given the very broad definition of confidential information - there was no necessary correlation between the two.
Was IlliquidX’s confidential information in the public domain?
That left the Court of Appeal to determine if there were sufficient grounds to interfere with the High Court’s finding that the confidential information relied on by IlliquidX was not in the public domain. It concluded that there weren’t. Whilst certain facts and general information were within the public domain (e.g. that Venezuelan debt was distressed and that it could be traded in spite of US sanctions), the specific confidential information relied on was not.
In reaching that conclusion, the court agreed with the High Court that sending marketing materials to a relatively large number of selected investors (about 200), with appropriate confidentiality markings, did not make them “generally accessible”.
Comment
This decision demonstrates that, absent drafting to the contrary, references to “public domain” in NDAs will generally be interpreted in line with the expression’s established meaning in the law of confidentiality. That meaning focusses on accessibility, rather than secrecy, with the burden being on those seeking to rely on the public domain exception to prove that the information is in the public domain. Parties wishing to broaden or change the intended meaning of “public domain” will need to use clear drafting to do so.
With thanks to Ioanna Antcheva, Catherine O'Callaghan, Callum Borg and Fraser Rodger for their contributions to this edition.
If you would like to discuss any of the above in more detail, please contact your relationship partner or email one of our IP team.
This material is provided for general information only. It does not constitute legal or other professional advice.