When AI Challenges Your Authorship: The Setback That Forces Us to Rethink Everything

View on LinkedIn

It was filed yesterday. The Author's infringement claim — one of those matters that seems settled even before it begins. The alleged infringer's conduct fits squarely within fraction I of Article 231. The evidence, impeccable. The registration, unimpeachable. No one saw what was coming next.

This morning, the expert opinion lands on the Institute's desk like a live grenade. The technical analysis, supported by AI-detection tools and an exhaustive comparative study of patterns, suggests that 99% of the work in question — the one the Author registered as his own — exhibits characteristics consistent with artificially generated content.

While the Author, who until a few hours ago had an airtight case, stammers explanations about "minor corrections" and "finishing touches," the opposing party does not merely defend itself — it counterattacks, filing a request for an administrative declaration of nullity of the registration.

The argument is devastating:

if the work was generated primarily by AI, it lacks the originality required to be protected under the Federal Copyright Law. There is no need to justify the unauthorized use of something that, technically, should never have been registered in the first place.

At first glance, the argument appears solid — especially when dealing with images generated by Midjourney or Stable Diffusion, or texts produced by ChatGPT. But the reality is more complex. What happens, for example, with software source code, where the extensive use of AI is already standard industry practice? Programmers around the world use Cursor, GitHub Copilot, Windsurf, and other AI tools to generate significant percentages of their code. Originality here does not necessarily reside in each individual line, but rather in the overall architecture, the selection of solutions, the integration of components, and the adaptation to specific needs.

If we apply the same "percentage generated by AI" criterion to software, we could be invalidating a substantial portion of current computer program registrations. Is that what the law intends? Or do we need to develop different standards of originality for different types of works?

But the true earthquake this case triggers within the Mexican copyright system goes beyond the validity of a single registration. We are witnessing a potential inversion of the foundational principles that have governed this area of law for decades. Copyright, by its very nature, arises the moment a work is fixed in a tangible medium, operating under a presumption of good faith that has been a cornerstone of the system.

Now, any work created after 2023 could face the burden of having to prove its creative process.

The question would no longer be "who copied whom?" but rather "can you demonstrate that you did not use AI?" How do we prove the absence of something? How do we demonstrate a creative process that, by its very nature, is rarely exhaustively documented?

Most authors do not save their initial sketches, their drafts, their failed attempts. A painter does not photograph every brushstroke, a writer does not document every rewrite, a musician does not record every improvisation. The creative process is frequently chaotic, intuitive, personal. Demanding its detailed documentation not only imposes an extraordinary evidentiary burden — it threatens to fundamentally transform the very nature of the creative act itself.

This inversion of the burden of proof could turn every copyright registration into a potential minefield, every infringement claim into an invitation to challenge the originality of the work from its inception. The copyright protection system, designed to foster and protect creativity, could paradoxically become its greatest obstacle.

Are we prepared for a world in which every author must become their own archivist, meticulously documenting each step of their creative process? Or do we need to fundamentally rethink how we understand and protect originality in an era where the line between human creativity and artificial assistance grows ever more blurred?