The Authorship Paradox in the Age of AI: Between Ownership and Creativity (November 2024)

View on LinkedIn

The question seems simple: if I create a work with the assistance of artificial intelligence, does it belong to me? From a patrimonial perspective, the initial answer is yes. The terms of service of platforms such as OpenAI and Anthropic confirm as much. However, the debate grows considerably more complex when we move from the patrimonial realm to the domain of copyright authorship.

Historically, authorship has been clear: if you use a paintbrush, a guitar, or a pen, the work belongs to the author. The same holds true for digital tools such as Adobe Illustrator or word processors. The disruption came with generative AI, and February 21, 2023 marked an inflection point when the U.S. Copyright Office [issued its decision in the "Zarya of the Dawn" case]( https://www.copyright.gov/docs/zarya-of-the-dawn.pdf ), initially rejecting registration of works generated by AI.

Yet the story did not end there. The Office itself refined its position in subsequent documents — first with a policy guidance on AI, and later with a more extensive document in the Federal Register. This evolution reflects the complexity of determining authorship in the digital age. (Artificial Intelligence and Copyright and Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence)

The crux of the matter lies in "human originality," a concept that demands careful analysis. Merely owning an AI tool or configuring an automated system is not enough. The key is the direct creative intervention of a human being in each individual work.

Consider a practical example: an automated system that downloads, classifies, and summarizes judicial precedents for a blog. Even if the system's creator owns both the blog and its content, the absence of direct human intervention in each individual summary breaks the chain of originality required for authorship. This phenomenon is more common than we might think: according to recent reports, 57% of content on the internet is generated or translated by AI.

The boundaries are clearer in certain cases. Unauthorized deepfakes, replicas that infringe registered trademarks, and plagiarism remain unlawful, regardless of the technological sophistication employed. The tool does not legitimize acts that are illegal by their very nature.

In Mexico, the current institutional stance is conservative. The National Copyright Institute (INDAUTOR) systematically rejects registration applications for works involving AI intervention — a position that has been upheld by the Specialized Intellectual Property Chamber. Nevertheless, this stance appears to be more of a temporary response than a definitive solution.

To better understand the nuances involved, let us examine the elements that determine human originality in the context of AI:

1. Creative Intervention: There must be meaningful human direction in the creative process.

2. Specificity: The intervention must be direct in each individual work, not merely in the general configuration of the system.

3. Transformation: The result must reflect human creative decisions, not simply the machine's own capabilities.

The analogy with traditional automated systems is instructive. When we program a coffee vending machine, we may define recipes and parameters, but each individual cup does not constitute an original work. Similarly, configuring an AI system to generate content automatically is not equivalent to being the author of each piece it produces.

The solution lies not in rejecting AI as a creative tool, but in establishing clear criteria for determining when sufficient human originality exists. We need a legal framework that recognizes both the potential of AI and the fundamental importance of human creativity.

The path forward requires a nuanced understanding of the intersection between technology and creativity. Creators seeking legal protection for AI-assisted works will need to document and demonstrate their specific creative intervention — beyond the mere operation of the tool.

At its core, the debate over authorship in the age of AI is not really about technology; it is about the very essence of human creativity.

As with any advanced tool, the difference lies not in its sophistication but in how the user applies their knowledge, judgment, and creative vision. Artificial intelligence represents a supertool that opens new possibilities for creation, but at the end of the day, that is precisely what it is: a tool.

Its true value resides in the hands of those who know how to use it. Those who today resist recognizing AI-assisted works are simply closing their minds to something they do not yet fully understand.

I look forward to seeing more attorneys litigating and forging new paths without fear — and, on the part of the authorities, a genuine understanding of what technology is and what their rulings actually entail.