As artificial intelligence continues to evolve, the legal and ethical questions surrounding intellectual property ownership and protection grow increasingly complex.
Artificial Intelligence (“AI”) has emerged as a disruptive force in various industries, including the realm of Intellectual Property (“IP”). AI’s impact on IP extends from automated IP searches to the generation of AI-created content, giving rise to complex legal and ethical questions around IP ownership and protection.
AI systems are increasingly producing creative works across various domains, such as music through Amper Music, translations with DeepL, poetry with Google’s Verse-by-Verse, and paintings via ArtBreeder. These creations often blur the line between human and AI-generated content. However, this does not guarantee that they will automatically receive copyright protection.
The primary legal concerns around AI-generated works involve establishing ownership of AI-created works, and determining the extent of copyright protection awarded by law to AI-created works.
As AI generates new forms of creative content, including music, literature, and art, it raises questions about IP ownership and protection. A central issue is whether AI-generated works can benefit from copyright protection.
For a work to be protected by copyright, it must be an artistic work and represent a concrete and original expression by an author. EU case law asserts that works can enjoy copyright protection if they represent “the author’s own intellectual creation”.
The originality standard contains two elements: (1) the work must not have been copied, and (2) it must present an intellectual creation. While AI can fulfill the first requirement, it struggles with the second, as it is inherently linked to a human author.
Established CJEU case law states that a work benefits from copyright protection only if the author made free, personal, and creative choices in the creation process, reflecting their personality. Economic investment alone does not suffice for copyright protection. Similar standards have been established by the US Supreme Court.
AI systems, however, operate differently from humans. They carry out deterministic processes based on the information provided and programmed, and cannot generate an outcome without prior data about similar objects. This contrasts with human creativity, which arises from internal stimuli and emotion, without the need for prior training. Consequently, AI systems cannot make the creative choices required for copyright protection.
In contrast, works created with AI assistance that still hold sufficient traces of human creativity can enjoy copyright protection. However, the degree of human guidance in AI creations is yet to be determined.
Creative choices depend on the type of AI system in question. Generally, there are three main types of systems:
(1) Step-by-step algorithms that use concrete if-then rules with little room for deviation.
(2) Rule-based algorithms that operate within margins provided by the programmer, allowing some room for AI operation.
(3) Machine-learning algorithms that learn from input data and can generate unknown style variations. Evolutionary algorithms fall under this category, as they optimize samples according to predefined criteria by autonomously selecting, mutating, and producing new samples that meet requirements.
The emergence of AI challenges traditional IP protection frameworks. In response, there is a growing need for legal frameworks that can accommodate AI’s unique aspects and its impact on IP, with some suggesting creating new IP rights to address AI.