The IP Brief - July 2024

Introduction

Welcome to our second 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 Court of Appeal’s interpretation of a minimum royalties clause in a trade mark licence between Virgin and Alaska Airlines;
the latest developments in the world of AI and IP, including the publication of the EU AI Act in the Official Journal (and what that means for IP) and the latest to-ing and fro-ing on the patentability of ANN-implemented inventions;
the UK Supreme Court’s decision in Lifestyle Equities v Ahmed, which looked at directors’ accessory liability and the remedy of an account of profits in the context of a trade mark infringement dispute;
the UK Patents Court’s refusal to grant Lenovo an interim injunction against Ericsson for infringement of one of Lenovo’s SEPs; and
two recent copyright decisions of the CJEU looking at communication to the public in the context of TV sets installed in: (i) hotel rooms and fitness areas; and (ii) rented apartment buildings.

TRADE MARKS

Virgin wins trade mark royalty dispute with Alaska Airlines

The Court of Appeal has considered the scope of a royalties clause in a trade mark licence agreement between two entities within the Virgin Group (“Virgin”) on the one hand, for whom Slaughter and May acted, and Alaska Airlines (“Alaska”, formerly Virgin America) on the other. Unanimously upholding the decision of the High Court, the Court of Appeal agreed that Alaska is obliged to pay an annual minimum royalty to Virgin for the right to use certain trade marks (the “Virgin Brand”) licensed to it by Virgin pursuant to the terms of a 2014 trade mark licence agreement (the “TMLA”), irrespective of whether Alaska actually makes any use of the Virgin Brand during any given year.

The Virgin Brand was first licensed to Virgin America under a 2005 agreement, pursuant to which Virgin America was: (i) subject to a strict ‘must-use’ obligation requiring it to use only the Virgin Brand when undertaking any licensed activities, and (ii) obliged to pay use-based royalties on all revenue derived from such licensed activities. As part of the process of obtaining approval for Virgin America to fly within the US from the US Department of Transportation (the “DOT”), and in particular to satisfy the DOT that Virgin America was not under the “actual control” of the UK-based (and thus foreign) Virgin entities, the parties agreed in 2007 to amend and restate the 2005 agreement to include the following new provision (Clause 3.7):

“Notwithstanding any other provision of this Licence nothing in this Licence shall prohibit the Licensee at any time during the Term from electing to perform the Licensed Activities or any other activities, including, but not limited to, operating flights, code sharing arrangements with any other airlines or entities, or operating flights between any points regardless of where such flights originate or terminate, without the payment of royalties, so long as the Licensee does not use the [Virgin Brand] while undertaking such activities”.

At that point in time there was no minimum royalty obligation contained in the agreement. However, ahead of the IPO of Virgin America in 2014 (and to compensate for, inter alia, the extension of the term of the agreement and the loss of certain rights and protections of Virgin which would occur as part of the IPO) the parties agreed to introduce a “Minimum Royalty” of c.$8m per annum. That minimum royalty was payable by Virgin America “[i]n consideration of the [rights in respect of the Virgin Brand] granted pursuant to [the TMLA]”. It was also made clear in a new Clause 8.6 that “[t]he Licensee’s obligation in respect of payment of royalties due to the Licensor in each financial year of the Licensee is to pay the greater of (a) [the relevant use-based royalty payable on revenue derived from use of the Virgin Brand], and (b) the Minimum Royalty payment applicable for that period…”. The wording of Clause 3.7 was not amended.

In 2016, Virgin America was acquired by a company in the same corporate group as Alaska; in 2018 it was merged with Alaska as a matter of US corporate law (such that Alaska became party to the TMLA in Virgin America’s place); and in mid-2019 Alaska ceased all use of the Virgin Brand. At the same time it ceased all royalty payments to Virgin, arguing that, in its opinion, the effect of Clause 3.7 (which it regarded as a ‘trumping’ provision) was that the Minimum Royalty was only payable when some use was made of the Virgin Brand.

In response, in December 2019 Virgin brought a claim seeking a declaration to the contrary and successfully obtained judgment in the High Court in February 2023. Alaska appealed and that appeal was heard in March of this year.

In a short judgment handed down in June, the Court of Appeal confirmed that the Minimum Royalty remained payable by Alaska irrespective of whether it made any use of the Virgin Brand. In brief, the court’s principal reasons for this were:

Alaska’s argument that Clause 3.7 ‘trumped’ all else in the TMLA was too simplistic - one first had to consider the entire TMLA and all its provisions so as to understand the overall meaning and effect of the contract and it was only once that meaning was ascertained that one could assess whether a particular clause actually conflicted with (and therefore was ‘trumped’ by) Clause 3.7.
Notwithstanding its name, the “Minimum Royalty” is not a ‘royalty’ in the ordinary use-based sense, but instead a minimum payment due in respect of the rights to use the Virgin Brand, regardless of whether Alaska actually made any such use. Interpreting the Minimum Royalty in this way meant that there was no conflict between Alaska’s Minimum Royalty obligations and Clause 3.7. In context, Clause 3.7 “entitles Alaska to conduct some (or indeed most) of its operations without using the Virgin Brand (derogating from the general requirement to use and promote the Virgin Brand in [C]lause 3.6) and not to pay royalties on those operations”, but does not entitle it to cease all use of the Virgin Brand and thereby avoid having to pay any consideration to Virgin whatsoever.
The 'factual matrix' and commercial considerations also supported this interpretation. On the former, it was clear in the court’s view that the Minimum Royalty was intended to provide protection to Virgin given the heightened risk post-IPO that a purchaser would take Virgin America private and cease use of the Virgin Brand (as Alaska in fact did); and the court considered it would be odd for the parties to “super-impose a minimum payment obligation in that context with the intention that it would be defeated by a complete de-brand, leaving Virgin with no recompense for licensing its rights”. As for commercial considerations, the court noted that there is a strong presumption that commercial parties do not intend to provide something for nothing, yet Alaska’s interpretation would enable it to hold and effectively “sterilise” the valuable Virgin Brand for up to 25 years (the term of the TMLA), without having to pay Virgin any consideration to do so. It also agreed with Virgin that Alaska’s interpretation would result in a “huge and unjustifiable difference” between a single use of the Virgin Brand in any part of Alaska’s operation (even by mistake), which would trigger the Minimum Royalty payment of nearly US$8m, and a complete de-brand, where nothing would be payable – an arbitrary distinction for which the court concluded there was no commercial or rational justification and which cannot have been intended by the parties.

Cases of this kind will always be heavily dependent on the specific wording used in the contract, but this decision serves as another reminder that clarity of drafting remains critically important.

The Court of Appeal’s guidance on “trumping” clauses is also worthy of note, with the court confirming that words giving primacy to one clause over another will be given effect, but only to the extent that there is actually an inconsistency or conflict between the relevant provisions.

Back to the top

AI AND IP

As ever, things continue to develop at a rapid pace in the AI space, with two important updates since our last edition of The IP Brief.

First and foremost, the final vote on the EU AI Act was passed on 21 May and the Act was published in the EU’s Official Journal on 12 July, clearing the way for the Act to come into effect on 1 August this year. Whilst not the focus of the Act, there are a number of copyright-related provisions in there which are particularly relevant for generative AI. These include obligations on providers of general-purpose AI models (which could include generative AI models) which have been placed on the EU market to:

put in place a policy to comply with EU copyright law (including any rights holders’ opt outs for text and data mining purposes); and
publish a “sufficiently detailed” summary of the content used to train the model, based on a template to be provided by the EU AI Office.

See our recent blog for more details. And see here and here for a broader summary of what the EU AI Act covers and what you can be doing now to prepare.

Separately, in May, the UK IPO released its updated guidelines for examining patent applications relating to AI (see our blog ). This followed a temporary period of suspension of the guidelines after the High Court’s decision in Emotional Perception AI Ltd v Comptroller General of Patents last year, where it was found that an AI invention relating to an artificial neural network (“ANN”) did not engage the exclusion from patentability for computer programs (see our blog). The updated guidelines reinforced the Emotional Perception High Court decision and temporarily offered reassurance to AI developers that patent applications for ANN implemented inventions wouldn’t face objections from the UK IPO based on the computer program exclusion.

In parallel, however, the UK IPO appealed the Emotional Perception decision. The Court of Appeal handed down its judgment on 19 July, overturning the first instance decision and confirming in the process that: an ANN is a computer; the weights and biases of an ANN (whether implemented in hardware or software form) are a computer program; and the computer program exclusion is therefore a relevant consideration when examining patent applications for ANN implemented inventions. Given these conclusions, the UK IPO has once again suspended its guidelines pending consideration of the Court of Appeal decision and will ultimately need to amend them for a second time. We will be publishing a blog on this latest decision shortly on The Lens.

Back to the top

GENERAL IP

UK Supreme Court considers directors’ accessory liability and account of profits

On 15 May 2024, the UK Supreme Court handed down its much anticipated decision in Lifestyle Equities v Ahmed. At its heart, this was a trade mark infringement case, but it gave rise to several important questions relating to accessory liability and the remedy of account of profits. In particular, the court looked at when a director might be liable as an accessory for causing a company to commit a strict liability offence such as trade mark infringement; and, where a director is found jointly liable, how the remedy of account of profits should be applied.

The court found, first and foremost, that there are no special rules for directors. To be jointly liable as an accessory for a tort, a director must know (or deliberately turn a blind eye to) the “essential facts which make the act unlawful” – even if the primary tort itself, such as trade mark infringement, requires no particular knowledge. The court was, however, keen to make it clear that a lack of knowledge of the law would not suffice to protect a director from liability.

It remains to be seen exactly what the “essential facts” will be for any particular tort and when a director may be regarded as having “turned a blind eye” to those facts. But, in an IP infringement context, it appears that the required knowledge (and thereby liability) will likely be found more readily in cases relating to counterfeit goods.

If accessory liability is found in a UK IP infringement context, and the successful claimant elects an account of profits as a remedy, the infringer will be required to return the profits they have gained from their wrongdoing to the claimant. However, the Supreme Court held in this case that an infringer can only be required to account for profits they themselves have made; they can’t be required to account for profits made by someone else (e.g. the company of which they are a director). It also found that, generally speaking, a proportion of salary or a loan made to a director should not be treated as “profit” for these purposes.

Back to the top

Patents

Patents Court refuses to grant PI in SEP proceedings

The Patents Court has, in a judgment handed down by Mrs Justice Bacon on 23 May 2024, refused an application by Lenovo for an interim injunction against the Ericsson group for infringement of one of Lenovo’s Standard Essential Patents (“SEPs”) covering aspects of 5G technology. Lenovo’s application was significant as there is no precedent for UK interim injunctions being granted in FRAND proceedings before the English courts.

Lenovo’s application arose in the context of the parties’ complex, two-way, multi-jurisdictional SEP dispute, which spans the UK, United States, Brazil, and Colombia. And it’s worth noting that, prior to this application, Ericsson itself had successfully obtained injunctions against Lenovo in Brazil and Colombia in respect of Lenovo’s alleged infringement of Ericsson’s own 5G SEPs in those jurisdictions – both of which were commercially significant markets for Lenovo (accounting for c.25% of Lenovo’s 2023 smartphone revenues).

The UK injunction that Lenovo sought was unusual because it was subject to a proviso that the injunction would not apply if Ericsson were to agree to one of its three “Preferred Alternatives”. Those alternatives consisted of various cross-licensing options tied to obligations which would restrict Ericsson from pursuing or enforcing injunctive relief against Lenovo in the UK or elsewhere (including enforcing the injunctions Ericsson had already obtained in Brazil and Colombia).

Applying the usual American Cyanamid test, the court agreed that there was a serious issue to be tried but the application fell down on the question of whether damages would be an adequate remedy for Lenovo. Generally speaking, in FRAND cases of this kind, where a SEP is infringed and the parties are not competitors, the patentee’s loss can usually be quantified as the sum that the patentee would have earned under a FRAND licence during the period in which it was enjoined. As a result, damages should be an adequate remedy. Lenovo, however, sought to argue that this case was different because its loss went beyond the sum that would have been earned under a FRAND licence. It submitted that, if the interim injunction was not granted, Ericsson would be allowed to continue to operate in the UK market while Lenovo would remain shut out from the Brazilian and Colombian markets under the injunctions Ericsson had already been granted in those jurisdictions. This in turn would enable Ericsson to exert significant commercial leverage on Lenovo to take a licence from Ericsson on terms it considered to be supra-FRAND in the interim period until a FRAND rate is determined by a court. Unfortunately for Lenovo, that argument was given short shrift by the Patents Court - the problem being that the damage Lenovo was seeking to rely on was not caused by Ericsson’s alleged infringement of the European (UK) patent in question; it was caused by Ericsson’s enforcement of its own patent rights in Brazil and Colombia. The High Court therefore concluded that damages would be an adequate remedy for Lenovo and the application for an interim injunction was refused.

While the UK has generally established itself as a key jurisdiction for settling global SEP disputes, it will be interesting to see whether this decision will prompt any changes in strategy from rights holders to focus at the outset on jurisdictions where they are more likely to obtain interim relief in order to create the commercial leverage that Lenovo were concerned about.

Back to the top

COPYRIGHT

CJEU considers communication to the public in context of TVs in hotel rooms and rented apartment buildings

The CJEU has considered the scope of a copyright holder’s exclusive right to control communications of its work to the public on a number of occasions. And it has recently done so again in two more cases, this time in the context of TV sets installed in: (i) hotel rooms and fitness areas; and (ii) rented apartment buildings.

In Citadines (Case C-723/22), the CJEU found that a hotel operator does carry out an act of communication to the public where it installs TV sets in hotel rooms and fitness areas and intentionally broadcasts copyright protected works (such as TV programmes) to its guests by retransmitting a signal to those TVs using the hotel’s own cable distribution network. The fact that the TV sets in question were not switched on by Citadines but by guests of the hotel operated by Citadines was deemed to be irrelevant. The CJEU also clarified that whilst the mere provision of physical facilities, such as TV sets, may not on its own amount to a communication to the public, if the hotel also distributes, by means of those TV sets, the signal transmitting the copyright protected work to guests staying in its rooms then there will be a communication to the public (irrespective of the technique used to transmit the signal). Whether or not Citadines will ultimately be found liable for copyright infringement, however, will depend on the scope of a licence for cable retransmission that Citadines had previously agreed with the German collective management associations – something which is a question for the German national courts to answer.

In GEMA (Case C-135/23), the CJEU concluded that the deliberate provision, by the operator of a rented apartment building, of TV sets equipped with an indoor antenna that pick up signals and enable broadcasts to be made will also amount to a communication to the public, provided that the tenants of those apartments can be regarded as a “new public” (or, indeed, a “public” at all). Whether or not the tenants in question could be regarded as a “new public” is a question for the German national court to determine, but will ultimately depend on the basis on which the apartments are let. If they are let on a short-term basis (e.g. as tourist accommodation), then the CJEU directed that those tenants should be classified as a new public. If, however, they are let to tenants who establish their residence there, then those tenants would not be regarded as a new public and there would be no copyright infringement. The reason for the distinction is twofold. Firstly, the case law of the CJEU has confirmed that the concept of “public” refers to “an indeterminate number of potential recipients” and implies a “fairly large number of people”. That number does not just refer to those who are able to access the relevant copyright works at the same time, but also how many people can access them in succession. As a result, determining whether the apartments in question (of which there are 18) are let on a short or long-term basis will have an impact on the number of people who may successively have access to the works in question - with more people having such access where the apartments are let on a short-term basis. Secondly, tenants in apartments that are let on a short-term basis may also constitute a “new” public, since those tenants, although within the catchment area of the relevant broadcast, could not enjoy the works being broadcast without the intervention of the building operator. In contrast, tenants who establish their residence in the apartments will be treated as having been taken into account by the copyright holder when they authorised the initial broadcast.

Both of these decisions highlight the broad scope of the right of communication to the public and serve as a reminder to consider carefully, when agreeing terms with rights holders, how the rights will be used and whether that is permitted. Whilst the English courts are no longer bound by CJEU decisions, we think it is likely that these decisions will be given some weight by the English courts, bearing in mind the Court of Appeal's previous comments on CJEU case-law relating to communication to the public in TuneIn v Warner Music.

Back to the top

Key contacts