No, AI-generated work cannot be copyrighted under current legal frameworks, as copyright protection strictly requires human authorship.
As artificial intelligence systems like Midjourney, ChatGPT, and Stable Diffusion transform the landscape of creative media, they have triggered a profound legal conundrum. While these tools can generate breathtaking images, write intricate software code, and compose symphonies in seconds, the law remains unyielding: creativity without a human pulse does not qualify for copyright ownership.
Here is a comprehensive breakdown of how global legal systems handle AI-generated content, the landmark cases shaping the future, and where the boundaries of human authorship are being drawn.
The Core Principle: Human Authorship
At the heart of global copyright law—including the frameworks of the United States Copyright Office (USCO), the European Union, and the UK—is the concept of the human creator.
In the United States, this dates back to landmark 19th-century legal precedents. In the famous 1884 case Burrow-Giles Lithographic Co. v. Sarony, the U.S. Supreme Court ruled that photographs could be copyrighted because they were visible expressions of human intellectual conception. Conversely, the law has consistently rejected non-human creators. This was famously demonstrated in the 2018 “Monkey Selfie” lawsuit (Naruto v. Slater), where federal courts ruled that a crested macaque could not own the copyright to a photo it accidentally snapped.
The USCO has explicitly mirrored this stance for artificial intelligence, stating it will register an original work of authorship only if the work was created by a human being.
Landmark Cases Setting the Precedent
The boundary between human and machine creativity is being actively litigated. Three major cases have defined the current legal landscape: Case / Entity Core Conflict Legal Outcome
Stephen Thaler & The Creativity Machine
Dr. Stephen Thaler tried to register an AI-generated artwork titled “A Recent Entrance to Paradise,” listing an AI system as the sole author.
Denied. The USCO rejected the application, and a federal judge upheld the decision, stating human authorship is a foundational requirement of copyright.
Kristina Kashtanova & Zarya of the Dawn
Author Kristina Kashtanova registered a comic book created using Midjourney AI text-to-image prompts.
Split Decision. The USCO revoked the copyright for the individual AI-generated images, ruling they weren’t human-authored. However, it allowed Kashtanova to keep the copyright for the arrangement of the images and the human-written text.
Jason Allen & Théâtre D’opéra Spatial
Jason Allen won a Colorado State Fair fine arts competition using a Midjourney-generated image, which he spent over 80 hours prompting and editing.
Denied. The USCO refused registration, stating that the visual output was ultimately generated by the machine, not human expression, despite the extensive prompting.
The “Prompt” Fallacy: Why Prompting Isn’t Authorship
A common argument made by AI creators is that the process of writing complex, multi-layered text prompts constitutes creative expression.
However, regulatory bodies view prompts similarly to giving instructions to a commissioned artist. If you tell a human painter to create a “futuristic city in the style of Van Gogh at sunset,” the painter owns the copyright to the resulting canvas, not you. Because the AI system determines the specific pixel placement, color blending, and composition, the machine acts as the generative force. The human prompter exercises “influence” rather than ultimate creative control. When Can an AI-Assisted Work Be Copyrighted?
While purely AI-generated outputs cannot be copyrighted, AI-assisted works can be protected under specific conditions. The key element is substantial human modification.
Compilation and Arrangement: If a human selects, arranges, and curates AI-generated elements in a highly creative and original way, the overall compilation can be copyrighted (as seen in the Zarya of the Dawn case).
Substantial Post-Processing: If an artist takes an AI-generated image or piece of text and significantly alters, draws over, or rewrites it using their own creative skills, the new human-authored layers can be protected.
AI as a Tool: Using AI as a mere technical aid—analogous to a camera’s autofocus, a digital paintbrush, or a spell-checker—does not invalidate copyright, provided the human remains the steering creative force behind the final output.
The Flip Side: Fair Use and Infringement
The debate over whether AI outputs can hold a copyright is only half the battle. The other half is whether AI companies infringed upon existing copyrights to train their models.
Major technology firms face massive class-action lawsuits from authors, visual artists, and software developers who argue that training AI models on billions of copyrighted works without consent or compensation constitutes theft. AI companies counter that this training falls under Fair Use, comparing the process to a human artist studying existing art to learn style and technique. The resolution of these cases will fundamentally reshape the financial and legal structures of the technology sector.
Looking Ahead
We are living through a transitional era of intellectual property law. As AI tools become more integrated into software, entertainment, and literature, distinguishing between pure machine output and genuine human curation will become increasingly difficult.
For now, the legal consensus remains clear: if you push a button and a machine does the heavy lifting of creation, the output belongs to the public domain. Copyright remains a uniquely human privilege, designed to protect the irreplaceable spark of human ingenuity.
Leave a Reply