Abby McDaniel
October 16, 2026
In news that will probably only make sense to the chronically online, popular Italian brainrot, Tung Tung Tung Sahur, is going to court. For those unaware, Italian brainrot is an internet trend consisting of dozens of artificial intelligence (AI) generated surrealist images that feature some combination of inanimate objects and animals, dubbed with Italian-esque sounding names [1]. To provide some examples, Chimpanzini Bananini is an AI chimpanzee growing out of a banana peel and Tung Tung Tung Sahur is an anthropomorphic stick holding a baseball bat [2]. The Italian brainrot trend primarily appeals to the young Generation Alpha, who have incorporated the phenomenona into games, merchandise, and videos. Meanwhile, most adults have dismissed the trend as nonsensical and largely irrelevant [1]. However, a recent lawsuit is bringing the issue of Italian brainrot to court in a case that will have massive implications on what content is applicable to US copyright law.
The lawsuit in question centers around an online game called Steal a Brainrot, made by an online creator known as SpyderSammy. Steal a Brainrot quickly gained popularity in late 2025 as a free to play game on the online multiplayer platform Roblox. The game has since garnered over 70 billion total visits with an average 125,000 active players at any given time [3]. In Steal a Brainrot, players collect various rarities of brainrots and then steal those brainrots from other players. Steal a Brainrot includes dozens of brainrot characters such as Chimpanzini Bananini and Tung Tung Tung Sahur [3].
However, the inclusion of Tung Tung Tung Sahur has proven to be legally problematic for Steal a Brainrot. In a claim filed to a California court, Tung Tung Tung Sahur’s creator, represented under a studio known as Mementum Labs, alleges that SpyderSammy has been illegally profiting off of his character. In response to the lawsuit, SpyderSammy temporarily removed Tung Tung Tung Sahur from the game [4]. Although SpyderSammy independently created Steal a Brainrot and coded it through Roblox’s game creation studio, he did not create the Italian brainrot characters his game features. Instead, the in-game characters were modeled to match the already internet famous brainrots, which Mementum Labs is claiming are copyrighted content [5].
The field of AI copyright law is in its formative days. Italian brainrot’s inclusion under copyright law is questionable, as the images are AI generated whilst copyright law has historically focused on only human generated works. Official US copyright law dictates that copyrightable works must be “original works of authorship,” wherein authorship must refer to a human being [6]. Case in point, the US Copyright Office ruled that a photo taken by a monkey or a mural painted by an elephant are unable to be copyrighted because the works are not human-made. One also cannot copyright a pattern that naturally occurred on a rock or copyright the shape of a piece of driftwood, because such items are the work of nature, not of any human [7]. Existing copyright law thus boils down to a complex question: How much human interaction is necessary to establish human authorship? In the case of AI, when humans prompt generative AI to create an image or essay in their place, is the human-AI interaction sufficient authorship for the establishment of a copyrightable product?
Until now, this question has not been definitively answered by any established precedent in copyright law. In the Italian brainrot case, the two sides of this debate are pitted directly against each other. Mementum Labs claims that the creation of Tung Tung Tung Sahur was based on a series of detailed, human-made prompts, so the character is thus protected under copyright law [4]. On the other hand, SpyderSammy’s studio, Do Big Studios, claims that AI generated media is the creation of a machine, not a human, and is thus ineligible for copyright [8]. Ultimately, the court’s decision in this case will shape the ability to file copyright claims on AI generated work such as photos, videos, and writing. The just outcome for this case can be determined by examining the historical context of copyright law.
When cameras were first invented, people protested the ability to copyright photographs, claiming that it was the camera creating the work, not the person. But the US Copyright Office took a different approach, outlining that a camera served as a “tool”, and photographs are copyrighted because of the human artistry required to compose the photo [6]. One might argue that this same lens could be used to view AI. AI could be considered a tool for human artistry and thus could be considered copyrightable [9]. However, this approach fails to acknowledge key differences between the methods of creation in AI and photography. When taking photos, the creator has full artistic control to frame whatever they want. The photo produced is under the full control of the photographer. The same cannot be said of AI work. Humans undoubtedly have a role in AI creation through prompts, but creators ultimately have very little control over what the AI creates. Oftentimes, AI will ignore parts of prompts and fill in unspecified information on its own accord. Furthermore, if a human pulls out a camera to take multiple photos, they can get many incredibly similar–if not the exact same–photos. If a person prompts AI with the exact same instructions multiple times, they will still end up with many vastly different creations [9]. The influence of human artistry is extremely reduced in AI because the creator has so little control over the final product.
Therefore, AI copyright law necessitates a more complex approach than photography copyright law. Copyright law is first and foremost defined by human authorship. AI generated work should only be able to be copyrighted when a human makes significant changes to its content, craft, or structure [6]. Merely prompting an AI to generate a photo or essay would not create a copyrightable product because the human authorship is too little. However, if a creator takes that AI generated essay and rewrites substantial parts of it or significantly changes syntax, then that essay should be copyrightable. Similarly, if a creator puts an AI generated photo into photoshop and edits it, then that photo should be copyrightable. In either case, the final line of what is copyrightable is if the final product is human touched rather than just pure AI.
Applying this mentality to the Steal a Brainrot case, SpyderSammy’s team is arguing that Tung Tung Tung Sahur was created merely through a series of prompts, which do not reflect enough human authorship to create the basis for copyright [8]. In opposition, Mementum Labs is arguing that Tung Tung Tung Sahur was developed through a series of prompts that explicitly and intentionally incorporated uniquely human traditions, namely related to Indonesian culture and Ramadan. Mementum Labs argues that the incorporation of human culture into the prompts suffices human authorship [4]. However, Mementum Lab’s claims are flawed in that they only consider the input and not the output. Yes, a human prompted AI to create Tung Tung Tung Sahur through his own unique idea. Yet when AI generated the image, the creator ultimately had minimal control over the output. Tung Tung Tung Sahur’s creator did not alter or in any way change the essentially random AI output and thus has no basis for human authorship under established copyright law. Hence, Mementum’s claims should be denied.
While the concept of Italian brainrot is undoubtedly juvenile and fatuous, the ruling in the Tung Tung Tung Sahur case will have much larger implications in the developing field of AI copyright. On the slight chance Mementum’s claims are upheld, much more AI created work could conceivably become copyrightable, even if humans have no significant influence on its production. Such a ruling would permanently alter the landscape of copyright law, allowing for vast expansion in AI generated music, books, code, photos, writings, and videos to flood the copyright market. Consequently, copyright law must hold firm in its human first precedent to protect the validity and usefulness of copyright, even if its largest challenger is currently an anthropomorphic stick.
Works Cited
[1] Zhan, Jennifer. “The Italian Brain Rot Ren-Ai-Ssance, Explained.” Vulture, May 29, 2025.
https://www.vulture.com/article/italian-brain-rot-ai-characters-explained.html.
[2] Sakshi. “Chimpanzini Bananini (Italian Brainrot).” Know Your Meme, March 31, 2025. https://knowyourmeme.com/memes/chimpanzini-bananini-italian-brainrot.
[3] Spyder, Brazilian. “[🌗] Steal a Brainrot.” Roblox, October 6, 2026. https://www.roblox.com/games/109983668079237/Steal-a-Brainrot.
[4] Spyder Games LLC and Speedy Simulator Gaming, LLC v. Mementum Lab. Case No. 5:25-cv-10248-NW. Case Management Statement and [Proposed] Order. U.S. District Court for the Northern District of California. Filed January 30, 2026. https://ia601904.us.archive.org/24/items/gov.uscourts.cand.460442/gov.uscourts.cand.460442.12.0.pdf
[5] Brumfiel, Geoff. “A Battle over ‘Italian Brainrot’ Could Shape Who Owns AI Art.” NPR, August 21, 2026. https://www.npr.org/2026/08/19/nx-s1-5867638/artificial-intelligence-brainrot-memes-copyright-spyder-tung-tung-sahur.
[6] U.S. Copyright Office. Compendium of U.S. Copyright Office Practices. 3rd ed. Chapter 300, “Copyrightable Authorship: What Can Be Registered.” Washington, DC: U.S. Copyright Office, 2021. https://www.copyright.gov/comp3/chap300/ch300-copyrightable-authorship.pdf
[7] “Monkey Selfie Lawsuit: Case, Ruling & 2026 Impact.” lawfold.com, May 12, 2026. https://lawfold.com/monkey-selfie-lawsuit/.
[8] Spyder Games LLC and Speedy Simulator Gaming, LLC v. Mementum Lab, Complaint, No. 5:25-cv-10248-NW (N.D. Cal. Nov. 26, 2025), https://ia601904.us.archive.org/24/items/gov.uscourts.cand.460442/gov.uscourts.cand.460442.1.0.pdf
[9] U.S. Copyright Office. Copyright and Artificial Intelligence, Part 2: Copyrightability. A Report of the Register of Copyrights. Washington, DC: U.S. Copyright Office, January 2025. https://www.copyright.gov/ai/Copyright-and-Artificial-Intelligence-Part-2-Copyrightability-Report.pdf