Today, an ordinary citizen with no formal art education can produce magnificent artworks in the style of Rembrandt or hyper-realistic photography in mere minutes simply by typing a short text prompt. So, who is the legal owner of this picture? The person who wrote the command? The company that coded the algorithm? The owners of the original images the algorithm was trained on? Or is there a masterless digital heap of data sitting in the "public domain"?
In this massive article, we delve deeply into this crisis that shakes the foundations of intellectual property law. Examining historical precedents, recent decisions by the US Copyright Office (USCO), the European Union's new AI Act vision, and Turkey's Law on Intellectual and Artistic Works (FSEK), we explore this dilemma across the axes of "Background," "Implications," and "Conclusions."
Can an algorithm's random arrangement of pixels be considered an artistic will?1. Background: The Anatomy of Copyright and Historical Precedents
To understand the AI crisis, we must first understand why modern copyright law exists. Intellectual property law was fundamentally invented to reward "human intellect and creativity," encouraging social development by economically protecting the artist.
1.1. "Sweat of the Brow" vs. "Modicum of Creativity"
Historically, two main doctrines have clashed regarding whether the law should protect a work. According to the "Sweat of the Brow" doctrine originating from English law, the labor and time spent creating a work (e.g., alphabetizing a massive telephone directory) were sufficient for copyright. However, this doctrine was struck down by the US Supreme Court in the famous Feist Publications v. Rural Telephone Service (1991) decision. The Court ruled that expending effort is not enough for copyright; a work must possess at least a "Modicum of Creativity." Alphabetizing a phone book was a mechanical task, not a creative choice.
This is exactly where the situation of people writing prompts to AI is debated today. Is entering the command "draw a red horse running under a blue sky" a spark of creativity, or is it a mechanical instruction, much like telling a taxi driver, "take me to Times Square"?
1.2. The Monkey Selfie Case (Naruto v. Slater)
Just before the AI lawsuits, the most bizarre "non-human creator" case the law faced was that of the famous macaque monkey, Naruto. Nature photographer David Slater left his camera in the jungle, and a monkey named Naruto took a selfie with it. The photo went viral globally. PETA (People for the Ethical Treatment of Animals) sued Slater, claiming the copyright belonged to the monkey.
Legal Precedent: Naruto v. Slater (2018)
The US 9th Circuit Court of Appeals ruled that copyright laws were written exclusively to protect "Human Authorship." Content created by an animal (or a natural phenomenon, wind, water), no matter how aesthetic, cannot be subject to copyright and immediately falls into the "Public Domain."
This decision formed an unshakable foundation for AI debates: Nothing without human will and individual creative intervention can be copyrighted.
2. The Global Legal Framework and Recent Rulings
Although there are slight nuances among the judicial systems of various countries today regarding the ownership of AI-generated images, a general consensus has begun to emerge globally.
2.1. US Copyright Office (USCO) and the "Zarya of the Dawn" Case
The most important milestone in this field is artist Kris Kashtanova's application to the US Copyright Office for the comic book *Zarya of the Dawn*. Kashtanova wrote the text of the comic book herself but generated all the visuals using Midjourney (AI). The USCO initially granted the copyright, but upon realizing the visuals were AI-generated, it revoked its decision.
The USCO's Ruling: The comic book's texts, page layout, and the way the visuals were curated are copyrightable because they contain Kashtanova's human creativity. However, the individual images produced by Midjourney hold no copyright. The USCO likened a user entering a prompt into an AI to a client giving instructions to a painter. You can say, "use red paint, draw a tree here," but the painter (i.e., the algorithm) decides where to apply the brushstrokes and how to refract the light. The outcome is unpredictable. There is a random machine gap between the image in the author's mind and the image produced by the AI. Therefore, pure AI outputs cannot be copyrighted.
2.2. Turkey: Law No. 5846 on Intellectual and Artistic Works (FSEK)
Article 1/B of FSEK, the heart of copyright in Turkey, clearly outlines the basic condition for a product to be considered a "work" (eser): "Any kind of intellectual and artistic product bearing the characteristic of its author..."
In Turkish law, "characteristic" (hususiyet) means the personality, mental effort, spirit, and creative choices of the person who created the work reflecting onto it. Since an algorithm does not possess a soul or personality, one cannot speak of the characteristic of a machine. Therefore, under current Turkish Supreme Court precedents and doctrine, an image created entirely by giving commands to an AI (without any human modification) cannot be considered a work under FSEK. Legally, these images are deemed Anonymous / Public Domain within Turkey's borders.
Legal Spectrum: AI Copyright Decision Tree
3. "Fair Use" and the Hidden Theft Crisis: Training Data
Discussing only the "output" dimension leaves the picture incomplete. The biggest apocalypse is breaking out on the "input" side. How do Midjourney or Stable Diffusion know Van Gogh's brushstrokes or the style of a famous living illustrator? Because these algorithms are trained by "web scraping" millions of copyrighted images available on the internet without permission (like the LAION-5B dataset).
Getty Images filed a massive lawsuit against Stability AI for allegedly using millions of its licensed photographs without permission to train its algorithm. In fact, distorted versions of the Getty Images watermark were clearly visible in some AI outputs. AI companies claim this is "machine reading," akin to a human going to a library, looking at thousands of pictures, and "taking inspiration" to draw their own, thus falling under "Fair Use." Authors, however, argue this is not "inspiration," but the direct copying of the digital weights of works to produce a "commercial imitation machine."
4. Implications: How Will Industries Transform?
This legal uncertainty and lack of copyright are having seismic impacts on creative industries:
- Advertising and Marketing Agencies: Agencies used to pay tens of thousands of dollars to stock photo sites (Shutterstock, etc.) for a campaign poster. Today, they generate these photos for free in seconds with AI. However, not owning the copyright is a problem. If a rival company copies the AI-generated visual exactly and uses it in their own ad, the agency has no right to object (regarding the visual). This creates a massive corporate identity risk for brands.
- Freelance Illustrators: The jobs of digital artists, especially those drawing backgrounds or creating character concepts in the gaming industry, are directly threatened. Companies prefer AI generations that, even if uncopyrightable, are "good enough" over human labor. The democratization of art (anyone can produce images) brings with it the devaluation of art.
- Film and Gaming Sectors: Companies like Marvel or Disney do not want their characters or scenes generated by AI, because any character they cannot copyright and license (and sell toys of) offers them no billion-dollar commercial return. Therefore, massive studios are compelled to invest in purely human-made works (for the copyright armor).
5. Conclusions: What Awaits Us in the Future?
It is clear that the current copyright regime (human-centric FSEK or USCO rules) cannot withstand this rapidly evolving technology for long. In a world where 90% of artworks are produced by machines in seconds and instantly deemed public domain, the internet will turn into a massive "uncopyrighted visual wasteland." So, what is the solution?
Legal scholars and policymakers are debating three main scenarios:
- The Zero Protection Scenario: The status quo continues. Machine output remains public domain. This could deter companies from using AI (if they desire copyright protection).
- Return to "Sweat of the Brow": If a person has spent days writing an AI prompt, tested 50 different words, and "curated" from hundreds of variations, this labor itself could be defined as a new "Work."
- Creation of a Sui Generis Right: AI-generated works are not granted full "Copyright" (because 70 years of protection is too long); instead, a brand new legal status called "AI Output Generator Right" (lasting 5 or 10 years) is invented, similar to the neighboring rights given to music producers, solely to prevent commercial copying.
Expert Opinion: Burhan Doğuş Ayparlar
Traditional legal approaches that equate the artificial intelligence revolution with the invention of the camera are operating under a grave misconception. With a camera, it was the human who adjusted the light, chose the framing, and decided on the exact "moment" to press the shutter; the camera was merely blind hardware recording that moment. However, Generative AI (Gen-AI) is not blind hardware; it is a decision-making partner capable of developing aesthetic judgment across billions of images.
In this context, the law's definition of the "Creator" is dead; the definition of the new era must be the "Curator / Director." A film director does not operate the camera personally, nor do they physically perform the actor's facial expressions; yet they orchestrate the entire process. The individual writing a text prompt to the AI is a digital director. Nevertheless, a simple and random command like "Draw a red car" does not meet the "characteristic" requirement necessary for copyright.
My proposed solution is this: It is imperative that our intellectual property laws (like FSEK) stop fighting the machine and define a new status. Pure copyright (the life of the author + 70 years) must absolutely not be granted to AI outputs; that is an insult to human creativity. However, to protect the commercial investments of agencies and producers, a new "Neighboring Right" named "AI Output Generator Right" (Sui Generis Right) must be invented, valid for only 10 years and aimed solely at preventing direct unauthorized commercial copying. This prevents cultural heritage (Public Domain) from falling under corporate monopolies, while providing businesses' investments and advertising identities with legal armor. Regarding training data (Input), establishing a mandatory "Extended Collective Licensing" pool to ensure artists receive "Fair Remuneration" is the only rational way out.
