News

Introduction

Generative artificial intelligence systems are trained on vast quantities of text, images, music and audiovisual content. For authors, artists, publishers, producers and collective management organisations, however, one central difficulty remains: how can they prove that a protected work was actually used to develop or deploy an AI model when the training data remain largely opaque?

A bill dated December 12, 2025, introduced by French Senator Laure Darcos, seeks to address this imbalance by establishing a presumption of use of cultural content by artificial intelligence providers. Adopted by the Senate on 8 April 2026, the bill was then transmitted to the National Assembly. It does not create a new intellectual property right: it changes the rules of evidence so that right holders can enforce their rights more effectively.

Why is it so difficult to prove the use of works by AI?

Model training relies on data that are rarely accessible

To establish copyright infringement, the right holder must, in principle, identify the work concerned, establish its rights and characterise the alleged acts of reproduction or exploitation. This becomes particularly difficult when content has been absorbed into vast datasets, assembled by several service providers and used to train a model whose internal workings are not public.

To learn more about the issues arising from the use of protected works to train AI systems, we invite you to read our previously published article.

Right holders may sometimes identify similarities in a generated output, prompt the reproduction of an element resembling their work or identify a reference to content in technical documentation. These elements do not, however, always make it possible to establish the origin of the training data with certainty.

The opt-out mechanism does not by itself resolve the evidentiary difficulty

Text and data mining consists of automatically analysing large quantities of digital content in order to extract information, trends or correlations. Article 4 of Directive (EU) 2019/790 authorises, subject to certain conditions, the reproductions and extractions necessary for such analysis where they concern content that has been lawfully accessed. For mining carried out for any purpose, however, this exception applies only if the right holders have not expressly reserved their rights. This ability to object to text and data mining is commonly referred to as an “opt-out”: the right holder expressly indicates, in particular by machine-readable means, that they do not consent to their content being used for this purpose.

Under French law, these rules are set out in Article L. 122-5-3 of the French Intellectual Property Code.

In practice, a reservation of rights is not always sufficient to protect the right holder. The right holder may object to the use of its work without being able to determine whether it was incorporated into a training dataset or whether its objection was respected. Without access to the technical information held by the provider, it may therefore remain difficult to prove unauthorised use.

How would the presumption of use of cultural content operate?

The right holder would have to provide an indication making the use plausible

The bill provides that subject matter protected by copyright or related rights would be presumed to have been used by an AI provider where an indication relating to the development or deployment of the system, or to the output it generates, makes that use plausible.

It would therefore not be an automatic presumption applicable to every work available online. The claimant would have to provide a sufficiently specific prima facie showing. A mere abstract assertion that a model must necessarily have been trained on cultural content should not be sufficient.

1. The right holder submits one or more credible indications.
2. The provider, which holds the technical information, may produce evidence to the contrary.

The mechanism thus places the burden of proof more closely on the party that actually possesses the information needed to determine the origin of the data and the conditions under which they were used.

The presumption would remain rebuttable

The AI provider could rebut the presumption by demonstrating, for example, that the work was not incorporated into the corpus, that it was used under a licence, that it came from a source covered by an authorisation or that the use validly fell within an exception.

The proposal does not therefore mean that every provider would automatically be considered an infringer. It would create an evidentiary tool, not strict liability.

In a favourable opinion issued on March 19, 2026, the French Conseil d’État considered that the mechanism could be reconciled with European Union law, subject in particular to using the more neutral term “use”, protecting trade secrets and limiting the mechanism to civil matters.

The presumption would encourage the use of licensing

The presumption would relate only to whether content was used; it would not, by itself, establish that the use was unlawful. A provider could therefore show that the work concerned was covered by a licence or another authorisation. This mechanism would encourage providers to document their sources and enter into agreements with right holders in order to secure the training of their models. The presumption would thus serve as a lever for accountability and negotiation, potentially fostering the development of individual or collective licensing arrangements.

What evidence could trigger the presumption?

The bill does not set out an exhaustive list. The assessment would therefore have to be carried out by the court on a case-by-case basis.

Evidence derived from generated outputs

• the repeated generation of elements substantially similar to a work;
• the reproduction of a passage, image, composition or details unlikely to result from mere coincidence;
• the appearance of signatures, watermarks, copyright notices or metadata associated with the original content;
• the system’s ability to reproduce a very precisely identified creative universe.

An isolated similarity would not necessarily constitute sufficient evidence. The analysis would notably have to distinguish the reproduction of protectable elements from the reproduction of a style, an idea, a genre or commonplace characteristics.

Evidence relating to the development of the model

• documentation published by the provider;
• a description of the datasets used;
• transparency reports;
• statements by service providers or researchers;
• licences acquired for certain categories of content;
• information disclosed during an expert investigation or a court-ordered evidentiary measure.

How would the presumption interact with European law?

It would not abolish the text and data mining exception

The presumption would not directly alter the conditions of the text and data mining exception. It would operate upstream of the legal analysis in order to determine whether the content was used. Once that use had been established or presumed, it would remain necessary to determine whether it was:

• authorised by a licence;
• covered by the text and data mining exception;
• carried out despite a valid reservation of rights;
• or constituted copyright infringement
• or an infringement of related rights.

The bill would therefore not automatically guarantee compensation. Its main effect would be to prevent a claim from failing before any examination of its merits because of a lack of access to evidence.

It would complement the transparency obligations under the AI Act

The AI Act requires providers of general-purpose AI models to put in place a policy designed to comply with European Union copyright law, including reservations of rights, and to publish a sufficiently detailed summary of the content used for training. Those transparency requirements do not, however, necessarily provide access to an exhaustive, work-by-work list of all the data incorporated.

The French presumption would thus serve a distinct function: facilitating the resolution of a civil dispute where the available information makes use plausible but does not yet make it possible to prove it directly.

Conclusion

The presumption of use of cultural content by artificial intelligence providers would not resolve every conflict between creation and AI. It would nevertheless provide a targeted response to one of the main obstacles encountered by right holders: the inability to prove technical facts under the exclusive control of their opponent.

Its effectiveness will depend on the definition of sufficient indications, the protection of trade secrets, its interaction with European law and the powers of the court to obtain reliable information. Cultural businesses already have an interest in formalising their reservations of rights and structuring the collection of evidence.

Dreyfus & Associés assists its clients in managing complex intellectual property cases, offering personalised advice and comprehensive operational support for the complete protection of intellectual property.

Dreyfus & Associés works in partnership with a global network of lawyers specialising in Intellectual Property.

FAQ

Can an AI system freely use every work available online?

The fact that content is available online does not mean that it is in the public domain. Its use may be covered by a licence or a statutory exception, or may require the right holder’s authorisation.

Is similarity between a generated image and a work sufficient?

Not necessarily. The similarity must be assessed in light of the original elements reproduced, the circumstances of generation and the other available evidence.

Can trade secrets prevent any disclosure of training data?

Trade secrets must be protected, but they do not necessarily preclude every evidentiary measure. Confidentiality mechanisms may allow a court or an expert to access certain information without making it public.

Why must AI providers strengthen the traceability of their data?

To rebut the presumption, they would have to be able to document the provenance of the data, the licences, filtering operations, the handling of opt-outs and the role of their subcontractors. Rigorous governance of training data would therefore make it easier to demonstrate lawful use or the absence of use of the work concerned.

What should a business do if its content appears to have been used?

It should preserve reproducible evidence (outputs, prompts, settings, date and model version), identify the works and rights concerned, review the relevant licences and reservations of rights and, with specialist counsel, determine the appropriate evidentiary measures and courses of action.

Introduction

Since April 14, 2026, the latest .de updates have changed the way .de domain names must be registered, transferred and administered. These changes, introduced by DENIC as part of its compliance with the European NIS2 Directive, are not merely technical adjustments for registrars. They directly affect trademark owners, legal departments, cybersecurity teams and domain name portfolio managers.

The central change is the move towards active verification of domain holder data. Information submitted when registering, transferring or updating a .de domain name may now be subject to a risk assessment. If the data appears incomplete, inconsistent or suspicious, a verification process may be triggered. Failure to resolve the issue within the applicable deadline may lead to suspension or even deletion of the domain name.

The significance of the .de updates for companies

The reform of .de domain name management forms part of a broader European cybersecurity framework. The NIS2 Directive aims to achieve a high common level of cybersecurity across the European Union, including for top-level domain registries and DNS service providers. Member States were required to transpose the NIS2 Directive by October 17, 2024, at the latest, and to implement the measures effective October 18, 2024.

For the .de namespace, this has resulted in stricter controls over the registration data associated with domain names under this extension. DENIC states that domain holders must provide accurate and up-to-date information, including the full name, legal form for legal entities, current postal address, email address and telephone number.

For companies, the issue is not merely administrative. A domain name may support a corporate website, an online store, email services, marketing campaigns, customer portals or brand protection strategies. A suspension can therefore have immediate operational and reputational consequences.

The .de domain name operations subject to verification

Since April 14, 2026, new .de registrations may be assessed at the time of order. Holder data is reviewed through a risk-based approach. If the information does not appear reliable, the registration may be blocked, quarantined or subject to further verification.

This approach is intended to reduce the use of false, unusable or manifestly inconsistent registration data. It is particularly relevant in the fight against phishing, fraudulent online stores, payment scams, trademark infringement and identity misuse.

Transfers and holder changes are also affected. A company wishing to move a .de domain name to a new provider or change the legal holder must therefore anticipate potential verification.

In practice, inconsistencies between internal records, registrar data, contractual documents and official company information should be avoided. A discrepancy in the company name, an outdated address or a missing legal form may be enough to delay or disrupt the transaction.

The .de updates also apply to contact updates on existing domain names. A change of address, email address, phone number or administrative contact may trigger a new assessment.

DENIC confirms that the obligation to maintain complete, accurate and verifiable data applies not only to newly registered domains, but also to existing domains, including those that have existed for many years.

The risks for trademark owners

The first risk is technical suspension, including through a serverHold status. In that situation, the domain name may stop resolving correctly. The website may become inaccessible, associated services may be disrupted and the company’s digital continuity may be affected.

For a trademark owner, the impact may be significant if the domain name is used for a .de website, local campaign, customer service platform, distributor portal or transactional service.

If the verification is not completed within the required timeframe, the process may ultimately lead to deletion of the domain name. Several technical providers have warned that unresolved verification requests may result in deactivation and later deletion if the data is not corrected.

This is particularly sensitive for strategic domain names. Deletion may create an opportunity for third-party registration, cybersquatting or portfolio disruption.

Companies with international portfolios often rely on several providers, several internal contacts and different management procedures. The new .de rules require stricter traceability.

Practical measures to prepare for DENIC verification requests

First, we recommend conducting a preventive audit of the .de portfolio. This audit must verify, at a minimum:

• the exact legal name of the holder;
• the legal form of the company;
• the complete and current postal address;
• the email address;
• the telephone number;
• consistency between registrar data and official corporate documents;
• the existence of an internal contact able to respond quickly to verification requests.

Companies using DomainBox should also monitor DomainUpdate poll messages in their API message queue. These messages may indicate that the registrant contact requires verification and specify the applicable deadline.

The practical risk often lies not in the verification itself, but in failing to detect the notification. A request received in an unmonitored technical environment may lead to an avoidable suspension.

Legal, IT and domain name teams should finally share a clear escalation process. When a verification request is received, the company should know who checks the data, who contacts the registrar, who validates supporting documents and who confirms that the issue has been resolved.

This risk-based approach is consistent with broader cybersecurity and data protection principles. The CNIL has emphasized that personal data security requires technical and organizational measures appropriate to the risks involved. This approach should also guide the governance of domain name portfolios.

Conclusion

The latest .de updates represent a significant change in the management of .de domain names. By strengthening holder data verification, DENIC has turned what was once an administrative formality into a genuine compliance and digital security issue.

Dreyfus Law Firm assists its clients in managing complex intellectual property cases, offering personalized advice and comprehensive operational support for the complete protection of intellectual property.

Dreyfus Law Firm works in partnership with a global network of attorneys specializing in Intellectual Property.

Q&A

Can a French company register a .de domain name?

Yes. The new rules do not prohibit foreign companies from registering a .de domain. Above all, they require that the registrant’s information be accurate, complete, and verifiable.

Should you avoid changing the contact information for a .de domain?

No. Outdated information poses a greater risk than a well-prepared update. You simply need to verify the information before making any changes.

Are defensive .de domain names affected?

Yes. Even if inactive or reserved defensively, a .de domain name must have reliable registrant information.

What should be done if a .de domain is managed by a former registrar?

You must identify the current registrar, verify the registered information, and, if necessary, arrange for a controlled transfer to your usual registrar.

Does the “serverHold” status mean the domain is lost?

No. It means the domain name is technically suspended. You must correct the information promptly to avoid subsequent deletion.

Do these new rules help combat online abuse?

Yes, indirectly. More reliable information can make it easier to identify abusive registrants, though it does not replace traditional trademark protection measures.

This publication is intended for general public guidance and to highlight issues. It is not intended to apply to specific circumstances or to constitute legal advice.

Introduction

Ahead of every major sporting event, including the FIFA World Cup, the Olympic Games and the UEFA European Championship, the same question lands on marketing directors’ desks: can a brand reference the event in its advertising without being an official sponsor? Where does the freedom to communicate end for a non-sponsor?

Between trademark protection, sports law, and unfair competition, companies must anticipate the risks attached to their commercial campaigns. Companies should therefore adopt a preventive approach to secure their marketing communications and limit the risks associated with their campaigns. Here is what businesses need to know about this practice, commonly referred to as “ambush marketing”.

Organizers’ exploitation rights: a protected monopoly {#organizers-exploitation-rights}

An intangible property right recognized by sports legislation

Under French law, sporting federations and event organizers hold an exploitation right over the competitions they organize, granting them a form of intangible property recognized by the French Sports Code. This right allows them to commercialize the event, notably through broadcasting rights, sponsorships and product licenses, and to finance their discipline. Courts, however, apply a restrictive interpretation: the monopoly protects the exploitation of the event itself, not every reference to the underlying sport.

Trademarks, mascots, slogans: FIFA’s intellectual property arsenal

Beyond this specific right, FIFA holds an extensive trademark portfolio covering tournament names, emblems, mascots, and host-city logos. This portfolio is subject to an active protection policy, including monitoring activities, online content takedown requests and, where necessary, legal proceedings. Reproducing or imitating these signs without authorization exposes advertisers to trademark infringement claims. The strategic management of a trademark portfolio is essential to prevent infringements of intellectual property rights.

To learn more about implementing a tailored trademark portfolio strategy, we invite you to read our previously published article.

Ambush marketing: a shifting legal boundary

Ambush by association

This form seeks to create, in the public’s mind, the illusion of an official link with the event through evocative symbols, colors, or slogans, without literally reproducing the protected mark. It is the hardest configuration to sanction, since advertisers carefully avoid any literal reuse of registered signs, making a straightforward infringement claim difficult to sustain.

Ambush by intrusion

Here, a company seeks physical or digital exposure near the event itself: distributing merchandise around stadiums, advertising within official “clean zones,” or running street activations timed with match schedules. This practice runs up against contractual and administrative venue regulations more than against trademark law as such.

Available legal grounds

In the absence of a single overarching statute, aggrieved organizers and sponsors typically rely on three grounds:

• trademark infringement, where a protected sign is reproduced;
• unfair competition and free-riding (parasitism), sanctioning a party that profits from another’s investment without bearing its cost;
• misleading commercial practices, where the communication misleads consumers into believing an official partnership exists.

Courts have therefore condemned promotional campaigns that use elements directly associated with a sporting event without the organizer’s authorization (see, for example, Paris Judicial Court, 12 February 2026, No. 23/15958). The financial penalties imposed vary depending on the nature of the infringement, the extent of the proven harm, and the investments made by the rights holders. These issues require a thorough legal assessment to evaluate the risks of trademark infringement, unfair competition, and parasitism (free-riding).

What can be done without authorization

Contrary to popular belief, non-sponsors retain considerable room to maneuver, provided certain guidelines are respected:

• referring to football or sport in generic terms (“this summer’s football event,” “the international competition”) without using protected names;
• using national colors or symbols not appropriated by a third party;
• building a campaign on creative association, without any logo or tournament reference;
• running lawful comparative advertising, within the applicable legal framework;
• ensuring influencer partnerships remain transparent about the absence of any official link with the organizer.

Some campaigns show that a brand can capitalize on sporting enthusiasm without using the distinctive signs of an official competition. A strategy built on sporting culture, the values associated with sport, or the visibility of individually contracted athletes can, under certain conditions, remain outside the scope of infringement.

How Can a Marketing Campaign Be Secured Ahead of a Sporting Event?

Before any campaign is launched, companies should review:

• the trademarks and signs used in the campaign;
• any direct or indirect references to the event;
• visuals generated using artificial intelligence;
• messages published by influencers;
• the risk of confusion with an official sponsor;
• the terms of use applicable to tickets, images, or sporting content.

The risks of overstepping the line

Poorly calibrated communication exposes a company to a combination of risks:

• civil liability: trademark infringement, unfair competition, or free-riding claims, with damages and court-ordered publication of the ruling;
• operational risk: immediate content takedowns on social platforms, under agreements between FIFA and networks such as Instagram or TikTok;
• reputational risk: a campaign perceived as unfair can durably damage brand image with the public and institutional partners alike.

A prior legal review of campaigns, particularly for international operations or those involving AI-generated content, has become an essential precaution. Dreyfus law firm also acts in pre-litigation and litigation strategies relating to infringements of intellectual property rights.

Conclusion :

Communicating around a major sporting event without being an official partner is possible, but it requires striking the right balance between creative marketing and respect for the organizers’ intellectual property rights. The absence of a direct use of a protected trademark or logo does not necessarily eliminate legal risk, particularly where a campaign may imply an official association, unfairly benefit from the organizer’s investments, or mislead the public.

At Dreyfus, we assist clients in navigating complex intellectual property matters by providing tailored legal advice and comprehensive operational support to ensure the effective protection and enforcement of their intellectual property rights.

Dreyfus law firm works in partnership with a global network of intellectual property attorneys, enabling us to support clients with cross-border IP matters and international enforcement strategies.

Nathalie Dreyfus with the support of the entire Dreyfus team

Q&A

Can a company use the name “World Cup 2026” in advertising without being a sponsor? No. This name is a registered trademark; unauthorized commercial use exposes the advertiser to an infringement claim.

How can a company ensure its campaign is legally compliant before launch? By conducting a legal review beforehand to verify that the campaign does not use protected trademarks or other protected signs, and does not suggest an official partnership or sponsorship.

Is using a player’s image enough to avoid the organizer’s intellectual property rights? No. Even where a company has entered into an agreement directly with a player, the campaign must comply with the player’s image rights, contractual obligations with their club, federation, or sponsors, as well as any rules applicable during the competition.

Can a company give away tickets to a sporting event as part of a promotional campaign? Ticket resale, transfer, and promotional use are often governed by the event’s ticketing terms and conditions. Companies wishing to offer tickets as prizes in a competition or promotional campaign should therefore verify the applicable contractual restrictions and, where necessary, obtain the organizer’s prior authorization

This publication is intended for general public guidance and to highlight issues. It is not intended to apply to specific circumstances or to constitute legal advice.

An abbreviation may create a likelihood of confusion with an earlier trademark where it is likely to be perceived by the public as its shortened version or as the sign of an economically linked undertaking. However, this assessment is never automatic: it depends on the similarity of the signs, the goods and services concerned, the distinctiveness of the earlier trademark and the perception of the relevant public.

The judgment of the General Court of the European Union of September 24, 2025 in Tamasu Butterfly Europa v EUIPO (T-326/24) provides a useful illustration: the removal of vowels from a word is not, in itself, sufficient to demonstrate that the public will perceive the new sign as an abbreviation of the earlier trademark.

Why can an abbreviation conflict with a trade trademark?

An abbreviation may function as an autonomous distinctive sign, a range name, a digital signature or a commercial identifier. Where it reproduces the distinctive core of an earlier trademark and covers similar goods or services, the consumer may believe that it is a variation, an affiliated service or a new presentation of the same undertaking.

• Under European Union law, Article 8(1)(b) of Regulation (EU) 2017/1001 allows the registration of a sign to be refused where there is a likelihood of confusion with an earlier trade trademark.
• Under French law, Article L. 713-2 of the Intellectual Property Code also covers the likelihood of association.

Tamasu Butterfly Europa v EUIPO: the facts, the decision and its significance

The facts

Tamasu Butterfly Europa GmbH opposed protection in the European Union of the word sign BTFY. It relied in particular on the earlier trademark BUTTERFLY, as well as on trade signs containing BUTTERFLY or BTY. In its view, BTFY had to be understood as a contracted form of BUTTERFLY obtained through “disemvowelling”, that is, by removing certain vowels.

The decision

The General Court dismissed the action. It acknowledged that disemvowelling is a technique known in certain digital and commercial uses. However, it refused to infer a general rule according to which any sequence of consonants would be spontaneously reconstructed by the consumer. In the present case, it had not been demonstrated that the relevant public would immediately perceive BTFY as an abbreviation of BUTTERFLY. The mere presence of common letters and the possibility of intellectually reconstructing the word were therefore insufficient to establish a likelihood of confusion.

The significance

The decision confirms that an abbreviation must be assessed as it is actually perceived, rather than on the basis of an ex post reconstruction. For the proprietor of the earlier trademark, evidence of market use therefore becomes decisive: commercial documents, press coverage, social media, online searches, distributor practices or evidence showing that the abbreviation is already associated with the full trademark. For the applicant, the judgment is a reminder that an abbreviated sign is not automatically available and must be subject to an appropriate clearance search.

How should the likelihood of confusion be assessed for an abbreviated trademark?

1. Examine the common distinctive elements

The number of common letters is not decisive. A short sequence may be highly distinctive if it concentrates the commercial identity of an arbitrary trademark. Conversely, elements such as “BIO”, “TECH”, “PRO” or “AI” will often have a weaker scope where they are customary or evocative in the sector concerned.

2. Compare the signs visually, phonetically and conceptually

The analysis concerns, in particular, the order of the letters, the initial elements, the length of the signs and their pronunciation. Conceptually, it is necessary to determine whether the public will see the abbreviation as an autonomous sign or as the natural shorthand of the earlier trademark.

3. Compare the goods, services and marketing channels

The same abbreviation may be acceptable in a remote market and problematic for identical or complementary goods, or goods sold through the same channels. Use in a domain name, an application, a marketplace or a social media account may also reinforce the association between the signs.

4. Identify the relevant public and its level of attention

Perception varies depending on whether the goods are aimed at the general public or at specialised professionals. A high level of attention may sometimes reduce the risk, but does not exclude it where the signs are similar and the goods are closely related.

The EUIPO Guidelines recall that these factors must be assessed globally and interdependently.

What steps should be taken before launching or challenging an abbreviation?

• Check earlier trademarks, as well as relevant trade names, domain names and unregistered uses.
• Determine whether the abbreviation reproduces the most distinctive element of a longer trademark.
• Anticipate the specific uses of the sign: logo, packaging, application, search engine optimisation, social media and advertising.
• Document the use of the abbreviation where it is already known to the public or distributors.
• Consider a separate filing where the abbreviation is intended to be used independently and on a lasting basis.

Conclusion: an abbreviation must be treated as a genuine trade trademark asset

An abbreviation may create a likelihood of confusion where it retains the distinctive identity of an earlier trademark and is used in a similar commercial context. However, it benefits from no automatic presumption, either as to its availability or as to its similarity with the full trademark. A concrete analysis, based on the perception of the public and evidence of use, remains essential.

Dreyfus & Associés assists its clients in managing complex intellectual property cases, offering personalized advice and comprehensive operational support for the complete protection of intellectual property.

Dreyfus & Associés works in partnership with a global network of attorneys specializing in Intellectual Property.

Q&A

Should the full trade trademark and its abbreviation be filed separately?

A separate filing is generally useful where the abbreviation is to be used independently on products, advertising materials, applications, domain names or social media. It also facilitates proof of rights and their enforcement.

Can a three- or four-letter abbreviation be protected?

Yes, provided that it is distinctive for the goods or services concerned. The shorter the sign, the more important it will be to examine whether it is commonplace, descriptive or already widely used.

Is removing vowels sufficient to avoid a conflict?

Disemvowelling does not automatically exclude similarity, but nor is it sufficient to establish it. Everything depends on whether or not the public reconstructs the full word.

Is an abbreviation used only internally protected?

Its internal use may constitute evidence, but it is generally insufficient to give it a trade trademark function in the eyes of the public. Consistent and identifiable external use strengthens its protection.

Is the risk the same in every country?

No, the risk is not the same in every country. Indeed, the pronunciation, meaning and familiarity of an abbreviation vary according to languages, commercial practices and the public concerned.

Introduction

The decision handed down by the French Supreme Court on February 18, 2026 (No. 23-23.681) forms part of a particularly significant line of case law in trademark law and contract law. It firmly reaffirms an essential rule that is often poorly anticipated by economic operators: the assignment of a trademark, even where it may be treated as a sale, does not automatically entail the transfer of the contracts governing its exploitation.

Through a dispute between a licensee-distributor and the purchaser of rights in a trademark, the French Supreme Court clarified the conditions under which licensing and distribution agreements may be enforced against an assignee. Although technical in appearance, this decision has considerable practical implications for any company exploiting intangible assets.

Analysis of the facts and contractual context

The case brought before the French Supreme Court arose in a complex economic and legal context, marked by several successive transactions affecting the company that owned the trademarks concerned, particularly in the context of insolvency and liquidation proceedings.

An economic operator benefited from a dual contractual relationship:

•  On the one hand, it held a trademark licence agreement, allowing it to exploit the distinctive signs owned by its contracting partner.
•  On the other hand, it participated in a selective distribution network, governed by a separate agreement setting out the conditions for marketing products under that trademark.

Following the transfer of the assets of the company that owned the trademarks, a new entity became the owner of the industrial property rights. The licensee-distributor then found that this new purchaser refused to supply it, which it interpreted as a wrongful termination of the contractual relationship. It therefore brought an action for contractual liability, seeking compensation for the loss allegedly suffered.

The licensee’s position: an extensive interpretation of the transfer of contracts

In support of its claim, the licensee developed a particularly structured argument based on an extensive interpretation of the mechanism for assigning trademarks.

According to the licensee, the assignment of trademark rights should be treated as a genuine sale of an intangible asset. On that basis, the trademark licence agreement should be regarded as a necessary accessory to that asset, in the same way as a lease attached to real estate. This analogy led to the argument that the assignee, having become the owner of the trademark, was required to guarantee the licensee peaceful enjoyment of it, which necessarily implied the continuation of the licence agreement.

The licensee also argued that the selective distribution agreement, which was in force at the time of the assignment, should follow the fate of the trademark, since it directly contributed to its economic exploitation.

This line of reasoning sought to establish that the transfer of the trademarks automatically entailed the transfer of the contractual relationships attached to them, thereby engaging the liability of the new purchaser in the event of termination.

The French Supreme Court’s decision: a restrictive approach to the transfer

The French Supreme Court dismissed the appeal and upheld the analysis adopted by the Bordeaux Court of Appeal, taking a particularly clear and rigorous position.

No automatic transfer of exploitation contracts

The Court recalled that the transfer of a business, including intellectual property rights such as trademarks, does not, unless expressly stipulated otherwise, entail the transfer of the contracts relating to their exploitation. This solution is consistent with the fundamental principle of the privity of contract, according to which a contract produces effects only between the parties that have consented to it.

In this case, the Court noted that neither the selective distribution agreement nor the trademark licence agreement had been expressly mentioned among the transferred items. Although the assignment deed referred to the company’s intangible assets, including trademarks and patents, it contained no reference to the continuation or assumption of these contracts.

This absence of an express provision was decisive: it excluded any automatic transfer of contractual obligations to the assignee.

Contractual interdependence between the trademark licence and selective distribution

The Court found that the parties had intended these two agreements to form an indivisible contractual whole, since the licence had been granted exclusively in consideration of the distributor’s approval within the selective distribution network. In other words, the right to exploit the trademark was directly conditional upon the licensee’s integration into that network.

In these circumstances, the absence of transfer of the selective distribution agreement necessarily entailed the absence of transfer of the trademark licence agreement. This contractual interdependence prevented any artificial separation of the obligations.

The Court concluded that the assignee, having not consented to the assumption of the distribution agreement, could not be regarded as a party to the licence agreement that depended on it.

No liability of the assignee

The direct consequence of this analysis is the complete absence of liability on the part of the assignee. Since it was not contractually bound to the licensee-distributor, it could not be subject either to a supply obligation or to any contractual obligation arising from the original agreements.

The refusal to supply the distributor therefore did not constitute a contractual fault, in the absence of any pre-existing legal relationship. The licensee’s claim for damages was accordingly dismissed in its entirety.

An essential clarification: the impact of insolvency proceedings

Finally, the Court specified that the successive insolvency and liquidation proceedings affecting the trademark owners had no bearing on the outcome. The disputed contracts had not been terminated, but they had not been transferred either. Their fate therefore remained governed by the ordinary principles of contract law.

Practical lessons for companies

This decision calls for increased vigilance in all transactions involving intangible assets. It serves as a reminder that the economic value of a trademark is not limited to its title of ownership, but depends closely on the contractual relationships that structure its exploitation.

For assignors, it is essential to identify precisely the contracts linked to the trademark and to organise their transfer expressly. Failing this, a substantial part of the value of the asset may disappear.

For purchasers, a thorough legal audit is indispensable. The absence of transfer of contracts may significantly limit the ability to exploit the trademark, particularly where there is no distribution network.

For licensees and distributors, this decision highlights the need to anticipate assignments contractually by providing mechanisms for the continuation or regulated termination of contractual relationships.

Legally securing trademark transfers

Current practice requires a structured and rigorous approach. It is essential to include precise clauses in assignment deeds relating to the transfer of contracts, clearly identifying the obligations assumed by the assignee.

Contractual drafting must also anticipate situations of interdependence or, conversely, provide for the autonomy of contracts in order to avoid uncertainty. The consent of co-contracting parties must be sought where necessary, particularly where intuitu personae (personal to the parties) clauses are involved.

Finally, a global analysis of the contractual ecosystem surrounding the trademark is essential in order to secure the continuity of its exploitation.

Conclusion

The decision of the French Supreme Court of February 18, 2026 (No. 23-23.681) establishes a clear principle: the assignment of a trademark does not, unless expressly stipulated, entail the transfer of the licensing and distribution agreements connected to it. This decision reinforces the need for precision in the drafting of assignment deeds and requires a strategic approach to intangible assets.

It serves as a reminder that a trademark is only one element of a broader framework in which contractual relationships play a decisive role in value creation.

Dreyfus & Associés assists its clients in managing complex intellectual property cases, offering personalized advice and comprehensive operational support for the complete protection of intellectual property.

Dreyfus & Associés works in partnership with a global network of attorneys specializing in Intellectual Property.

Nathalie Dreyfus with the support of the entire Dreyfus team

Q&A

1. What is the difference between a trademark assignment and a trademark licence?

A trademark assignment entails the transfer of ownership of the distinctive sign to a third party, who becomes the new holder of the rights. By contrast, a trademark licence is merely an authorisation to exploit the trademark: the owner retains ownership of the trademark but allows a third party to use it under contractually defined conditions. This distinction is fundamental because it determines the rights and obligations of the parties, as well as their liability.

2.What are the risks of poorly drafted trademark licence agreements?

Insufficiently precise drafting may have significant consequences, such as:

•  uncertainty as to the scope of the rights granted;
•  difficulties in exploiting or controlling the trademark;
•  disputes relating to the duration or termination of the agreement;
•  a loss of economic value of the trademark;
•  or even challenges to the validity of the agreement.

3.Must the assignment of a trademark be published in order to be enforceable against third parties?

Yes. Under French law, the assignment of a trademark must be recorded in the National Trademark Register maintained by the INPI in order to be enforceable against third parties. Failing this, even if the assignment is valid between the parties, it cannot be invoked against third parties, which may weaken the assignee’s position, particularly in litigation or in the context of competition.

4.Can a trademark licensee object to the assignment of the trademark?

In principle, no. The owner of a trademark is free to dispose of its property right and may assign the trademark without obtaining the licensee’s consent. However, exceptions may apply where the licence agreement contains an approval or prior information clause, or where the licence is concluded intuitu personae, meaning in consideration of the identity of the initial owner.

5.Can a trademark be assigned independently of a business?

Yes. Unlike certain elements of a business, a trademark is an autonomous intangible asset that may be assigned independently of the other elements of the undertaking. However, such a transaction must be handled with caution, particularly where the trademark is closely linked to a specific activity or customer base, in order to avoid any risk of deception or depreciation.
This publication is intended to provide general guidance and highlight certain issues. It is not intended to apply to specific situations or to constitute legal advice.

Introduction

Quantum computing is now moving beyond the field of research and gradually entering certain industrial applications, particularly through cloud-based access to specialised computing resources. This development requires traditional IT contracts to be adapted in order to anticipate the risks specific to quantum technologies, including technological uncertainty, rapid obsolescence, performance issues, intellectual property, data protection and regulatory developments.

Identifying the contractual model: QaaS or a dedicated quantum computer

Even before negotiating the technical clauses, the parties must identify the access model chosen. Contracts relating to quantum computing do not all cover the same reality. They may involve two distinct models:

• The first is Quantum Computing as a Service, or QaaS, under which the client remotely accesses quantum processors hosted by a service provider, in a manner similar to cloud computing.
• The second is based on the provision of a dedicated quantum computer, sometimes installed on the client’s premises or at a designated site.

This distinction is decisive, since the clauses to be included will vary depending on the model chosen. In a QaaS model, the contract must notably regulate access to the platform, service availability, execution queues, usage-based billing, data location and reversibility. For a dedicated computer, the contract should instead specify the hardware specifications, installation constraints, acceptance tests, maintenance, insurance, transfer of risk and conditions for returning the system.

These elements determine the service provider’s undertaking: depending on the degree of control over the technology, this may amount to a best-efforts obligation, an enhanced one or, more rarely, an obligation to achieve a specific result.

Allocating the technological risks specific to quantum computing

For many uses, quantum computing remains only partially mature, still in the process of industrialisation. This partial maturity creates specific risks that the contract must allocate clearly and fairly between the service provider and the client.

Regulating the service provider’s liability

Under French law, limitation of liability clauses are permitted, provided that they do not deprive the debtor’s essential obligation of its substance, in accordance with Article 1170 of the French Civil Code. In a contract relating to quantum technologies, drafting such clauses therefore requires a clear distinction to be drawn between:

• Limitations inherent in the technology, such as quantum noise, decoherence, measurement errors or instability of results; and
• Breaches attributable to the service provider, such as poor algorithm design, defective execution or insufficient integration into the client’s information system.

The service provider cannot guarantee the complete absence of computational errors where these result from physical constraints specific to quantum computing. However, it must be able to demonstrate the means implemented to limit, correct or reduce their effects. This demonstration requires the contract to organise the evidentiary framework for establishing a breach, particularly in QaaS environments where it may be difficult to determine whether an incorrect result stems from a defect in the service or from the normal uncertainty of the technology. The contract may therefore provide for traceability obligations: execution logs, timestamped metrics, incident reports and technical audit procedures.

Providing for technological obsolescence

Quantum solutions evolve quickly: a tool, algorithm or access to infrastructure may become less relevant within a few months. The contract must therefore organise the evolution of the solution over time, without relying solely on the statutory regime of hardship. It may include a technological update clause, a right to renegotiate or terminate in the event of significant obsolescence.

Where the contract concerns a dedicated quantum computer, it must also anticipate the gradual deterioration of performance: ageing components, drift in error rates, regular recalibrations or software updates. Minimum performance thresholds may be defined, together with an obligation for the service provider to intervene, a price reduction or a right of termination if those thresholds are no longer met.

Protecting intellectual property rights and data

Quantum contracts raise issues in the field of intellectual property. Quantum algorithms, circuit compilation methods, optimisation techniques and results generated during performance of the contract may constitute strategic assets for the parties.

Determining ownership of algorithms and results

The service is not always limited to the use of an existing tool: it may lead to the creation or improvement of algorithms, optimisation methods, circuits or technically exploitable results. The contract must therefore specify what falls within the elements already owned by each party, what is developed specifically for the client and what the service provider may, where applicable, reuse after completion of the project.

This distinction is essential, since these elements are not necessarily intended to be subject to the same legal regime: some may remain the property of the service provider, while others may be assigned to the client or only be subject to a licence of use.

Finally, the contract must regulate the service provider’s reuse of certain elements resulting from the project, particularly aggregated data or results, where they are used to improve its models, algorithms or services.

Regulating confidentiality and data protection

The data processed under quantum contracts may be sensitive, strategic or subject to specific regulatory obligations. This vigilance does not only concern personal data: industrial, financial, scientific, pharmaceutical or defence-related data may have significant strategic value, even where they do not fall within the scope of the GDPR. The contract must therefore regulate their location, the service provider’s access conditions, restrictions on use, segregation measures, as well as deletion, return and reversibility procedures.

Particular attention must also be paid to the so-called “harvest now, decrypt later” threat, which consists of collecting encrypted data today in order to decrypt them later using quantum computing capabilities. Where the sensitivity of the data so requires, the contract may organise a gradual migration towards post-quantum cryptography standards, in accordance with ANSSI recommendations.

Anticipating regulatory developments and dispute resolution

The legal framework applicable to quantum technologies remains largely indirect and fragmented. Existing texts do not always refer to quantum technologies as such, but may apply depending on the concrete use of the technology: personal data processing, use of cloud infrastructure, cybersecurity or migration to post-quantum cryptography.

The contract must therefore include compliance clauses adapted to the context of the project. In terms of export control, the parties must notably take into account Regulation (EU) 2021/821 on dual-use items and technologies, as well as, where the project has an international dimension, the US EAR regime and, in certain circumstances, ITAR. The contract must then allocate compliance responsibilities and provide for the necessary representations, warranties and cooperation obligations.

The sector-specific regulations applicable to the client must also be taken into account, particularly in the finance, healthcare, defence or critical infrastructure sectors. Depending on the circumstances, obligations arising from DORA (Regulation (EU)2022/2554), the NIS 2 Directive (Directive (EU) 2022/2555) or equivalent frameworks may need to be passed on to the service provider, particularly in relation to security, continuity, audits and incident notification.

Providing for a regulatory adaptation clause

As these contracts must be capable of adapting to changes in the applicable standards, the contract must provide for an adaptation mechanism allowing the parties to identify relevant regulatory developments, assess their technical, economic and operational impact, and then renegotiate the affected provisions. This clause may organise a reciprocal duty to inform, a renegotiation period, the allocation of compliance costs and the consequences of persistent disagreement.

Article 1195 of the French Civil Code may indeed allow renegotiation in the event of an unforeseeable change in circumstances making performance excessively onerous. However, in such an evolving field, it is preferable to organise adaptation scenarios contractually in order to avoid uncertainty regarding the conditions for continuing, suspending or terminating the contract.

Providing for an appropriate dispute resolution mechanism

Disputes relating to quantum technologies may raise complex technical questions, particularly where they concern the origin of a performance defect, the reliability of results, the attribution of an error or the compliance of the solution with the contractual specifications. It is therefore useful to provide, in advance, for a mechanism capable of clarifying the technical aspects of the

dispute before any contentious proceedings.

The contract may thus organise recourse to an independent expert with suitable expertise, responsible for analysing the causes of the difficulty encountered and issuing a technical opinion. This prior expert assessment may be supplemented by mediation or arbitration where the dispute persists.

Conclusion

Quantum computing contracts require specific drafting adapted to technologies that remain only partially mature and are evolving rapidly.

Their legal security rests on four key pillars: identifying the appropriate contractual model, allocating technological risks, protecting intellectual property and data, and anticipating regulatory developments and disputes.

The objective is to manage technological uncertainty while preserving the value of the investment. When negotiated upstream, these clauses help secure the legal position of the parties.

Dreyfus law firm  assists its clients in managing complex intellectual property cases, offering personalized advice and comprehensive operational support for the complete protection of intellectual property.

Dreyfus law firm works in partnership with a global network of attorneys specializing in Intellectual property.

Nathalie Dreyfus with the support of the entire Dreyfus team.

FAQ

1. Should a company already enter into contracts for quantum projects while the technology remains emerging?
Yes, precisely because the technology remains emerging. The contract makes it possible to frame expectations, limit areas of uncertainty and avoid the project being based solely on technical or commercial promises. It is particularly useful during proof-of-concept phases, where the parties must know whether they are testing a hypothesis, developing a reusable asset or preparing for future industrial deployment.

2. How can a contract be prevented from becoming overly marketing-oriented?
The contract must convert commercial promises into verifiable commitments. General expressions such as “revolutionary solution”, “quantum advantage” or “significant optimisation” should be replaced by objective criteria: the scope of the service, comparison methodology, testing conditions, performance indicators and expected deliverables.

3. Must the client provide its own data to test the solution?
Not necessarily. Where the client’s data is sensitive, it may be preferable to begin with fictitious, synthetic, anonymised or representative data. If real data is necessary, the contract must specify its scope, the purposes for which it may be used, the security measures to be implemented and the conditions for deletion or return at the end of the project.

4. How can dependence on a single provider be avoided?
The contract must anticipate the risk of technological lock-in. This requires provisions on usable output formats, sufficient documentation, a reversibility clause, migration conditions and, where possible, an architecture designed to avoid excessive dependence on a single technology or platform.

5. Should the contract provide for an enhanced duty to advise?
Yes, particularly where the client does not have equivalent technical expertise. The service provider must inform the client of the limits of the technology, the risks of failure, the conditions necessary for the success of the project and the possible alternatives. This duty to advise may be decisive where the client relies on the service provider’s expertise when deciding whether to invest.

The purpose of this publication is to provide general guidance to the public and to highlight certain issues. It is not intended to apply to particular situations or to constitute legal advice.