Numerique
In a world where artificial intelligence is revolutionizing artistic creation, an unprecedented legal battle is currently playing out in international courts. Generative AI technologies such as Midjourney, DALL-E and ChatGPT now produce works of impressive qual
Reading time:
7 min
In a world where artificial intelligence is revolutionizing artistic creation, an unprecedented legal battle is currently playing out in international courts. Generative AI technologies such as Midjourney, DALL-E and ChatGPT now produce works of impressive quality, blurring the traditional boundaries of copyright.
This technological revolution raises fundamental questions: who owns the rights to an image created by an AI? Can artists whose works were used to train these systems claim compensation? Is current legislation suited to these new challenges?
If you wish to engage a digital law attorney, contact me!
The litigation surrounding generative AI has reached considerable proportions since the high-profile case pitting three visual artists against Stability AI, Midjourney and DeviantArt. These creators accuse the companies of having used their works without consent to train their algorithms, amounting in their view to a massive infringement of intellectual property. The complaint filed before a California court in January 2023 paved the way for a series of similar proceedings, notably the one brought by Getty Images against Stability AI.
Case law is only just beginning to take shape on these unprecedented questions, with early rulings that diverge across jurisdictions. In the United States, the Copyright Office adopted a restrictive position by refusing to grant copyright protection to works generated entirely by AI, finding that they lack the necessary "human authorship". By contrast, certain works partially created with the assistance of AI have been able to obtain limited protection, marking a subtle distinction that will shape the future of artistic creation.
The implications of these early court decisions are considerable for the creative industry. From animation studios to design agencies and content publishers, every player in the creative ecosystem must now navigate a zone of legal uncertainty. This situation creates an environment conducive to disputes, where the line between legitimate inspiration and automated infringement is becoming increasingly blurred.
The question of originality lies at the heart of the current legal debate. Generative AI systems work by analyzing millions of existing works to extract patterns and features that they then recombine. This process raises fundamental questions: can a work generated by AI be considered original within the meaning of copyright law? Or is it merely a sophisticated form of derivation from pre-existing works?
Faced with these complex legal uncertainties, calling on a digital law attorney becomes essential to navigate this legal maze. A professional qualified in this emerging field can not only identify the vulnerabilities specific to your situation, but also design preventive strategies suited to the rapid evolution of case law. For companies developing or operating creative AI technologies, consulting a digital law attorney represents a prudent step that allows rigorous compliance verification processes to be established and ongoing legal monitoring to be put in place. These legal experts play a crucial role in interpreting the provisions of the European AI Act regarding the transparency of AI-generated content, while assisting creators and companies in drafting contracts suited to this new technological reality. Their in-depth understanding of the technical mechanisms underlying AI systems enables them to anticipate risk areas that generalist lawyers might fail to identify.
Another complex legal aspect concerns the moral rights of artists, which are particularly important in the French legal system. Moral rights, which include in particular the right of attribution and the right to respect for the integrity of the work, are perpetual and inalienable. How do they apply when an AI generates an image in the style of a living or deceased artist, without directly reproducing their works? This question currently divides legal scholarship and will likely be the subject of clarification through case law in the coming years.
Let's discuss your needs for 15 minutes!
A particularly interesting phenomenon concerns the emergence of "prompt engineers", these professionals who specialize in drafting precise instructions to obtain specific results from AI systems. Their work raises a fascinating legal question: can the instructions given to an AI be protected by copyright? Some courts are beginning to recognize a form of protectable creativity in the crafting of particularly sophisticated prompts, thereby opening up new territory for intellectual property law.
The economic consequences of these decisions are considerable. Entire business models are currently being built around AI content generation, with massive investment in the sector. Legal uncertainty constitutes a major risk for these players, all the more so since damages in copyright infringement cases can reach astronomical amounts, particularly in the American legal system where statutory damages allow for fixed compensation per infringed work.
The civil and criminal liability of developers and users of generative AI remains largely undefined. If an AI generates a work that turns out to be an infringement of a protected creation, who bears responsibility? The company that developed the algorithm? The user who formulated the request? Or perhaps the person who selected and published the result? These crucial questions will shape the future of innovation in the field of creative AI and determine the level of risk associated with the use of these technologies.
Faced with this uncertain legal landscape, human creators are developing a variety of defensive strategies. Some artists now embed invisible watermarks in their digital works in an attempt to disrupt AI training. Others turn to more restrictive creative licenses or join platforms that allow them to explicitly control whether their creations may be used to train algorithms.
On the side of businesses using generative AI, legal risk management has become a major concern. Emerging practices include regular audits of training datasets, obtaining explicit licenses for the works used, and the implementation of traceability systems making it possible to identify the origin of the elements composing the generated creations. Some companies are also developing mechanisms for the automatic remuneration of creators whose works contributed significantly to the outputs of their algorithms.
The major technology platforms are adopting varied approaches to this legal challenge. Some, like Adobe with its Firefly system, claim to train their models on royalty-free or explicitly licensed works. Others prioritize technical efficiency, even if it means facing potential lawsuits later on. These differences in approach reflect distinct corporate cultures and risk strategies, but also a differing anticipation of how the legal framework will evolve.
I want reliable legal documents!
The legal decisions made today will durably shape the creative ecosystem of tomorrow. Beyond the technical questions of intellectual property, it is the very value of human creation that is at stake. European legislators appear to be moving toward a model demanding greater transparency about the use of protected works in AI training and an explicit acknowledgment of the artificial origin of generated content.
This period of legal transition represents both a challenge and an opportunity. Creators who adapt to this new reality, by understanding the legal mechanisms at play and working with these technologies rather than against them, will likely be best positioned to thrive in this new paradigm. At the same time, companies that anticipate legislative and case-law developments by adopting ethical and transparent practices now will be able to limit their exposure to litigation risks while fully benefiting from the creative potential of AI.
The development of an ethical standard for the use of generative AI in artistic creation could play a crucial role in resolving these tensions. Initiatives such as origin certification of training data and the mandatory labeling of AI-generated content are emerging as potential solutions to reconcile technological innovation with respect for the rights of human creators.
The intersection between generative artificial intelligence and copyright constitutes one of the most dynamic frontiers in contemporary legal evolution. The tension between technological innovation and the protection of human creators will not find a simple resolution, but rather a gradual balance through a series of judicial and legislative decisions. What is certain is that we are witnessing the birth of a new chapter in intellectual property law, where human and artificial creativity will have to learn to coexist within a legal framework suited to the realities of the 21st century. The support of an attorney specialized in artificial intelligence law and an internet law expert will be essential to navigate this new legal era.
To learn more
The question remains largely open. Technologies such as Midjourney, DALL-E or ChatGPT blur the boundaries of copyright, which has traditionally been founded on human creation. As things stand, a work generated purely by AI struggles to be protected, for lack of an identifiable human author.
This is one of the core issues of the legal battle. Creators accuse AI companies of having used their works to train their systems without authorization or compensation. The litigation, such as the case pitting artists against Stability AI and Midjourney, illustrates this issue.
The dispute concerns in particular the use of works to train AIs without the creators' consent. The case pitting visual artists against Stability AI, Midjourney and DeviantArt illustrates this clash between human creators and AI systems over the question of rights.
The question is open. Copyright law was designed around human creation and struggles to grasp works generated by AI. This mismatch fuels the debates on the necessary evolution of the legal framework to meet the challenges of generative AI.
Not without caution. AI-generated content is not always protectable, which limits its appropriation, and it may infringe pre-existing rights. The company must verify the absence of any infringement and contractually frame the use of such content.
This is a central question of the dispute. Creators claim compensation for the use of their works in AI training. The legal framework is evolving, but the answer remains uncertain and is the subject of major litigation before international courts.
It is necessary to verify that the generated content does not infringe existing rights, to document the human involvement in order to seek protection, and to contractually frame the ownership of the productions. A cautious approach limits the risks of infringement and legal uncertainty.
A digital law attorney helps to assess the risks associated with generative AI, to secure the use of content, and to anticipate the evolution of copyright law. This support is invaluable in a field caught up in a full-fledged legal battle.
Still have questions?
Our team is available!
Have a question?

Ressources
Aller plus loin