[Guest post] Aesthetic theft or shared style? Originality in the social media age

The IPKat has received and is pleased to host the following guest contribution by Katfriend Adanna Onah on a recent decision of the US Court of Appeals for the Ninth Circuit in Woodland v Hill. Here is what Adanna writes:

Aesthetic theft or shared style? Originality in the social media age

by Adanna Onah

The Ninth Circuit decision in Woodland v Hill offers additional insight into the challenges of applying copyright law principles in digital contexts. The decision reiterates that copyright in photographs only bars ‘selection’ and ‘arrangement’ of elements, not the bare idea of a pose.

Background

Rodney Woodland is an artist, figure model, and photographer. He has a particular fondness for tastefully-posed, semi-nude portraits. His Instagram feed features a curated series of images posted between 2018 and 2021, mostly garnering modest engagement (i.e., a few dozen likes).

Montero Lamar Hill, known as Lil Nas X, is a prominent pop musician. Between March and October 2021, Hill uploaded eight promotional photographs, also semi-nude. According to Woodland, Hill’s images were so similar to his photographs that they crossed from inspiration into infringement.

Thus, in June 2022, Woodland sued Hill for copyright infringement, alleging that Hill had copied his images. The District Court was not persuaded and found no plausible evidence that Hill had access to Woodland’s posts, and ruled that the photographs were not substantially similar. Woodland appealed, but the Ninth Circuit affirmed the District Court’s decision.

Selection and arrangement: Does it provide clarity or confusion?

The Ninth Circuit framed its reasoning around the Rentmeester test, which states that only the photographer’s unique ‘selection’ and ‘arrangement’ of elements (poses, lighting, camera angle, background, etc.) are protected by copyright law. Additionally, quoting from Feist¸ the court noted that another artist may copy any single element so long as the overall arrangement is not identical:

Thus, “[a] second photographer is free to borrow any of the individual elements featured in a copyrighted photograph, ‘so long as the competing work does not feature the same selection and arrangement’ of those elements.” Id. at 1120…

Woodland argued that his and Hill’s photos have too much in common, but the court disagreed, stating that the overall compositions differ enough that no infringement could be established. The court also conceded that there is no “well-defined standard” for when selection-arrangement similarity is substantial, highlighting that each case turns on its own facts.

Tropes, scènes à faire, and creative genres

For creators in trope-heavy genres like epic fantasy or action pictures, the ruling restates a long-held copyright law principle that, with the idea/expression dichotomy and the scènes à faire doctrine, stock elements are unprotectable ideas. Only a distinct creative arrangement or twist on these tropes can gain protection. In the case, the court treated Woodland’s common place poses (e.g., a reclining male figure or an outstretched arm) as nothing more than familiar building blocks.

The court repeatedly noted that Woodland’s claims rested on the reproduction of ideas (i.e., a similar naked model vibe) rather than any original expression. Citing the classic “ordinary observer” test, the court stressed that an ordinary viewer must see the same selection and arrangement in both works. In the court’s words, “Ultimately, the photos’ selection and arrangement of elements must be similar enough that ‘the ordinary observer, unless he set out to detect the disparities, would be disposed to overlook them.’”

The court observed that Hill’s photos “share few similarities” with Woodland’s, but a generic pose or mood is not enough to establish copyright infringement. For example, in analysing Woodland’s Morning Fog and Hill’s Montero, the court initially stated that each image depicts a nude Black man with a bright light obscuring the groin area and set in a blue-toned backdrop. But once zoomed in, the differences started to become apparent: In Woodland’s, the light source creates shadows and mystery, the subject’s head is turned away, with the face obscured. In Hill’s image, the light is direct and reveals glistening skin, and one of the models stares confidently at the camera. Accordingly, the court concluded that, although both images riff off the same concept, the compositions differ too much to support an infringement claim.

Algorithmic discovery and digital copying

Woodland, furthermore, attempted a shortcut in proving copying. He argued that because both he and Hill post niche content (i.e., artistic nude Black male modeling), Instagram’s recommendation algorithm must have shown Woodland’s photos to Hill. This argument was rejected, and it was held that merely using the same social media platform is insufficient to prove that a defendant had access to a plaintiff’s works. Rather, a plaintiff needs actual evidence demonstrating that the defendant viewed the work (a “chain of events”) or that the work was widely disseminated. The court went further to state that Instagram’s guidelines emphasise that its algorithms on suggested pages are mainly influenced by who users follow and what they like, not a mysterious content-matching rule as Woodland suggested.

Conclusion

The principles applied in this case are not particularly novel but are nonetheless important. The case deals with Instagram, but the logic transfers to other content platforms such as TikTok and YouTube with the GRWM sequences, clean girl aesthetic, and other viral edits that no doubt may have originated from one creator.

The decision may also lend credence to the argument often employed: publicly posted content is fair game for machine learning. Visual content, once posted online, is often stripped of exclusivity unless it exhibits and preserves a specific, original configuration, and even then, only if copying is provable. In essence, originality, while essential in US copyright law, does not win a case alone; proving that copying occurred is also important. Therefore, it is not enough for a plaintiff to show that two works are similar. They must demonstrate that the defendant had access to the original work and that the similarities are substantial enough to infer copying rather than independent creation. 

This is not far from the argument often advanced by AI developers: if the content is online, publicly visible, and largely comprised of standard repeated elements, then its inclusion in training data does not violate copyright law, especially if the output is not substantially similar. Of course, Woodland v Hill dealt with human copying, not automated ingestion, and therefore may be viewed differently by a court. While human copying involves conscious intent, machine copying raises more complex questions about whether ingestion and reproduction by an algorithm amount to the same legal standard of copying. Nonetheless, the case’s relevance lies in the court’s emphasis on the need to prove actual copying or near identical similarity between the disputed photographs. The court’s insistence on observable appropriation of original expression as opposed to assumed infringement offers a useful parallel as courts grapple with the question of AI training and AI-generated content.

However, the court may have missed an opportunity for clarification. The decision anchored protection on a standard bordering on near-identical sequencing. In fairness, the opinion also reflects a court struggling to reconcile Feist’s originality doctrine with the exponential growth of derivative content. The vast, repetitive nature of social media compresses individual aesthetic expression into something ephemeral and ubiquitously accessible.

So, what is the advice for creators? Keep posting, but it is unlikely that copyright law will protect individual elements of your aesthetic or signature style.