by on March 4, 2026
Earlier this week1, the U.S. Supreme Court (SCOTUS) denied certiorari (cert) in Thaler v. Perlmutter. Thaler contended that an image — generated by a Large Language Model (LLM)-backed Artificial Intelligence (AI) — deserved copyright registration. Since the U.S. Copyright Office refused to grant the registration, Thaler appealed to the U.S. District Court for the District of Columbia (DC Circuit). That Court affirmed the Copyright Office's decision. SCOTUS' denial of “cert” means they will not hear the case. Strictly speaking, this denial does not affirm the DC Circuit Court's ruling, but it does mean the DC Circuit decision stands.
Many in the Free and Open Source Software (FOSS) community raised concerns about the impact on copyleft — and even FOSS in general. TL;DR: Don't Panic! — this case is extremely limited in scope.
First, a proviso: this case is about copyright of an artistic image, not software. Copyright law — and the legal precedents around it — differ widely for different types of creative works. Analysis of the copyrightability of works of software varies in notable ways. Therefore, do not to assume that analysis for images apply broadly to software.
Second, while the decision is “published” 2, there are also many other cases related to LLMs and AI currently pending throughout the U.S. Courts. Courts and laws always lag behind technological advancement. Indeed, this is precisely why copyleft was invented: as a mechanism to achieve with existing laws and precedents what we could not accomplish in the legislature. Forty-one years after copyleft's invention, we still do not have a federal law that mandates software right to repair!
Third, the Court found that a registration was not valid (at this time) if the work's sole author is a computer program. Thaler (who was both (one of) the author(s) of that computer program and its user) repeatedly waived any claim to consider Thaler's own copyright in the LLM-backed AI prompting process. Thaler also did not argue any copyright interest in the LLM-backed AI system itself were subject of the registration. So, this decision does not evaluate any creative expression by (a) the author(s) of the prompts themselves, (b) copyrights held in the LLM, its weights, generation, curation, or its user interface, and (c) copyrights held in underlying works in the LLM training data.
Thaler's original registration was the root cause of this substantial narrowing because the registration contended that the AI system itself was the author of the image. This case only considers a copyright registration where the sole “author” is identified as a specific computer program. Thaler stipulated that the work was generated solely through prompts and no human modified the work thereafter. As such, even if the other districts begin citing this case regularly, and even if many districts decide it applies to software without further consideration of the difference in the types of works, such precedent causes no disaster for FOSS.
Admittedly, some LLM-backed generative AI agents can be merely prompted to create a work of software from scratch that has some transient utility. However, the most common workflow in using these agents (at least in FOSS development) is as follows:
Furthermore, dicta 3 — appearing in the DC Circuit ruling — supports a conclusion that the human actions on that third step would constitute creative expression — affixed in tangible medium — suitable for copyright registration. Indeed, their ruling states:
First, the human authorship requirement does not prohibit copyrighting work that was made by or with the assistance of artificial intelligence. The rule requires only that the author of that work be a human being — the person who created, operated, or used artificial intelligence — and not the machine itself. — (Thaler v. Perlmutter, 130 F.4ᵗʰ 1039, 1049 (D.C. Cir. 2025))
The Court also indicated these other issues are for a future time in another case. The DC Circuit readily admits that their ruling applies only to the state of AI systems at the time of writing4. Again quoting from their ruling:
Of course, the [Thaler's AI] Machine does not represent the limits of human technical ingenuity when it comes to artificial intelligence. Humans at some point might produce creative non–humans … Science fiction is replete with examples of creative machines that far exceed the capacities of current generative artificial intelligence. For example, Star Trek’s Data might be worse than ChatGPT at writing poetry, but Data's intelligence is comparable to that of a human being. See Star Trek: The Next Generation: “Schism” (Paramount television broadcast Oct. 19, 1992) (“Felis catus is your taxonomic nomenclature, an endothermic quadruped, carnivorous by nature”). There will be time enough for Congress and the Copyright Office to tackle those issues when they arise. — (Thaler, 130 F.4ᵗʰ at 1050)
[ As always, SFC is not a law firm, IANAL, and TINLA. ]
1 I — and my colleagues at SFC — acknowledge that SCOTUS made other decisions recently regarding an array of important social justice causes. Since SCOTUS' decision to deny cert in this particular case is so closely related to my work, I'm writing about it. However, all of us at SFC acknowledge that our community is reeling from other recent decisions.
2 In this context, “published” is a term of art that lawyers use to describe a case that the publishing Court (in this case, the DC Circuit) felt was important enough to “share officially and formally” with other Courts. While (in a precedent-based legal system) any Court can cite an unpublished case from another Court, published cases are much more likely to be cited than unpublished ones.
3 “Dicta” is explanatory language found in a court's decision that isn't necessary to the court's conclusion. . Dicta isn't precedential but it can be persuasive.
4 Note that the DC Circuit issued their ruling in March 2025. It is not uncommon for SCOTUS to delay for a year (or more) before issuing a ruling to grant or deny cert.
Please email any comments on this entry to info@sfconservancy.org.