In Brief:
- Current UAE Law Does Not Recognise AI as an Author: Under the UAE Copyright Law, while natural or legal persons can be considered owners of a copyright work, the concept of the creation of a work of copyright rests on a natural person.
- Human Creativity and Input Remains Central to Copyright: The article explores how authorship depends on the degree of human involvement - mere use of AI is not enough; creative input, direction, and refinement by a human - the ‘human touch’ - is essential.
- Case by Case Analysis: Pending specific legislative updates (both in the UAE and elsewhere) to reflect the emerging doctrines on the intersection between copyright and works generated with AI, it is recommended that the question of whether copyright can subsist in a work created with the use of AI is considered by an analysis of the human input into the creation of the work, and mindful that wholly AI-created elements will not fall under the protection of copyright.
1. Introduction
The rapid emergence of low-cost (and often free), easily accessible and easy to use Artificial Intelligence (AI) platforms has given an unprecedented ability for anyone to unleash their inner Picasso and create what is sometimes (but not always) very impressive works of art.
When one is lacking any artistic ability, as with the author, a few minutes on Bing Image Creator creates the 2025 masterpiece of the lawyer artist:
Source: Created using Bing Image Creator
With the significant advancements in AI and its applications in various domains, numerous legal challenges have emerged.
The current ‘hot topic’ with AI and copyright is on the use of copyrighted materials as ‘training data’ for deep learning-based generative artificial intelligence models and whether this is an infringement of copyright.
This brief article does not aim to exhaustively address the intricacies and complexities of this evolving subject but rather sheds light on key issues and general principles. More detailed articles exploring the nuances of different types of works will follow in due course.
For the purposes of this article, a separate core question for a work generated through the use of AI is whether the resulting work can be a copyright work, and what is the current legal position in the UAE on this matter.
Through a brief historical review, this article discusses:
-
- The role of technology in creating a work protected by copyright.
- Can a non-human be considered the ‘author’ of a copyright work?
- Can a work generated with AI be considered a copyright work?
As will be seen, the interplay between copyright, authorship and technology is not a new concept.
2. What is copyright?
To place this article in context, a brief overview of copyright is necessary.
The World Intellectual Property Organisation (WIPO) defines copyright as:
“Copyright (or author’s right) is a legal term used to describe the rights that creators have over their literary and artistic works. Works covered by copyright range from books, music, paintings, sculpture, and films, to computer programs, databases, advertisements, maps, and technical drawings.”
The US Copyright Office helpfully adds:
“Copyright is a type of intellectual property that protects original works of authorship as soon as an author fixes the work in a tangible form of expression.”
In the UAE, Federal Law No. 38 of 2021 on Copyrights and Neighboring Rights (the UAE Copyright Law) defines a “work” in Article 1 as:
“Any innovative creation in literature, arts, or science, regardless of its type, mode of expression, significance, or purpose”.
3. What rights are afforded by copyright?
A key feature of copyright is that it grants the owner of a copyright work (usually the author or otherwise the entity to whom the author has assigned the ownership of the copyright) the right to decide the basis on which the work can or cannot be used by others.
Under most copyright laws, the following exclusive rights are granted:
- The right to reproduce (copy) the work. Unauthorised reproduction or copying is the most common type of copyright dispute.
- The right to create derivative works by making changes to the underlying work. These changes can include editing, modifying, or translating the work or, in the case of a written work, an adaption into a different form (such as a movie).
- The right to control the manner of distribution of the work. This can include by selling, lending or licensing the work.
- The right to control the public display of a work.
- The right to control the public performance of a work.
These ‘economic rights’ can be valuable, or grow to be valuable, and so the question of whether copyright subsists in the particular work, in order to exploit these economic rights, is crucial.
4. Who is the author of a copyright work?
As a general position, the creator of a copyright work is considered to be the author of the work, provided of course that it is original and has not been copied from elsewhere. This is a key element of the debate on the use of ‘training data’ and whether copying has occurred.
Underlying this is the commonly held position that only a human can be the author of a copyright work.
In the UAE, as is the case under copyright laws around the world, an author of a copyright work is considered to be a ‘person’. Article 1 of the UAE Copyright Law defines an author as:
“A person who creates a Work, whose name is mentioned on a Work or to whom a Work is ascribed for being its owner, unless otherwise established. An Author of a Work also includes any person who publishes a Work without a name or under penname or in any other way; provided that the Author's true identity is not a matter of doubt. Should a doubt exist, the Work's publisher or producer, whether a natural or legal person, shall be deemed a representative of the Author in exercising his rights until the Author's true identity is recognised.”
But what is a ‘person’? The UAE Copyright Law defines a person as “a natural or legal person” so while a ‘legal person’ (such as a company) may own a copyright work (and which is common position from an IP asset ownership framework for companies) a ‘legal person’ cannot be regarded as an “author” in the strict legal sense.
This article takes the position that authorship inherently requires human contribution – a ‘human touch’ – and while ownership of a copyright work can vest in a non-human, can a non-human be considered an ‘author’ of a copyright work? Does it make a difference where AI is involved in the creative process?
During Dubai AI Week 2025, these questions are the focus of this article.
5. The Human Touch
How much human input is required for a resulting work, created through AI, to be protected by copyright?
This question, arising as it does from a technological development (in this case, AI) is not unique. More specifically, the application of copyright principles has had to take account of various technological developments at different points in history.
Prior to the emergence of technology which allowed the capturing of an image (via the heliograph in 1816 by Joseph Nicéphore Niépce and the daguerreotype by Louis Daguerre in 1839), the depiction of a person, place or thing was through the use of paint, charcoal or other means onto a medium, usually canvas or paper. There could be no question that the artist was the creator (author) of the resulting work and, provided that copying has not occurred, that copyright subsisted in that work.
The rapid development of photographic technology over the mid-to late 19th century was accompanied by an explosion of popularity in the cabinet card – a mounted photograph approximately 6 x 4 inches in size and displayed in the home on a table or cabinet (hence the name). Cabinet cards also operated as an advertisement for the photographer with details of the photographer and studio appearing on the back of the cabinet card.
Commonly used for personal portraits of family and friends, cabinet cards featuring famous actors, actresses and other celebrities were extremely popular from the 1870s to 1900 and were often collected and sold – the beginning of ‘celebrity’ merchandise and memorabilia. This provided the battlefield for whether the use of technology (a camera) to generate an image would result in a work protected by copyright.
Napoléon Sarony was an American lithographer and highly popular 19th century portrait photographer based in New York and was especially popular with celebrities to photograph cabinet cards. In January 1882, Sarony took the following photo (Wilde Image No. 18) of Oscar Wilde:
Source: https://www.oscarwildeinamerica.org/
At the time of the photograph, Oscar Wilde was better known for being a wit and a dandy on the London society circuit and his fame from the “The Picture of Dorian Gray" (1891) and other plays was still years in the future.
The Burrow-Giles Lithographic Company (Burrow-Giles) was a company that specialised in prints and lithographs, and increasingly cabinet cards. Many of these cabinet cards were commissioned by department stores, and where the cabinet cards contained the contact details or other advertisements of the department stores.
Burrow-Giles produced 30,000 cabinet cards for the New York department store Ehrich Bros. and 100,000 cabinet cards for Mandel Bros. - a Chicago department store. In both instances, Burrow-Giles used Wilde Image No. 18 without permission:
Source: https://www.oscarwildeinamerica.org/
Sarony commenced proceedings against Burrow-Giles for copyright infringement. While it was not disputed by Burrow-Giles that Sarony has operated a camera to take the photo, Burrow-Giles argued that a photograph could not be protected by copyright since it was not authored by a person but rather was “a reproduction, on paper, of the exact features of some natural object, or of some person”.
In response, Sarony asserted that he was the author of the photograph, and therefore of a copyright work, since he had envisaged and implemented the arrangement of the elements of the photograph, including the position of Wilde and his pose, the outfit worn, the type and location of other props, and deciding on the light for the photograph.
The Federal Southern District of New York found in favour of Sarony, which was further affirmed by the U.S. Circuit Court for the Southern District of New York, and subsequently affirmed by the Supreme Court of the United States. In its decision of 17 March 1884, the Supreme Court, taking into account the input of Sarony in the photographic arrangement as well as the act of taking the photograph, found:
“this photograph to be an original work of art, the product of plaintiff's intellectual invention, of which plaintiff is the author, and of a class of inventions for which the Constitution intended that Congress should secure to him the exclusive right to use, publish, and sell”.
Having established that a photograph could be protected by copyright, a further technological development in the form of the Kodak Brownie camera (launched in February, 1900)
Source: https://www.eastman.org/collections-online
not only saw a decline in the popularity of cabinet cards, since the public now had access to an affordable (USD 1 dollar) means of candid photographs, but granted the ability for public to themselves be authors of copyright works in the form of photographs which could be easily taken, developed, reproduced, and commercialised.
A similar access to cheap and advanced technology emerged in the 2020s with the rise of AI (as discussed below).
6. Monkey Business
Over a century later and 16,000 km away, David Slater (a British photographer) was taking photographs of a troop monkeys in North Sulawesi, Indonesia in July 2011.
Having set up a camera on a tripod and made settings, the equipment drew the curiosity of the monkeys, including Naruto (a Celebes crested black macaque), who took the following ‘monkey selfie’:
Source: Wikipedia
This subsequently went viral around the world and appeared on various websites over the course of 2011, including on Wikipedia.
Efforts by Mr Slater to commercialise the photograph included requests to hosts of the photograph to take the photograph down. Wikimedia (the non–profit organisation behind Wikipedia) refused on the basis that the photograph was in the public domain since it was the creation of a non-human and, as a result, copyright did not subsist in the photograph.
The Slater/Wikimedia dispute ended up before the US Copyright Office where, on 21 August 2014, it was held that the photograph was not protected by copyright since:
“only works created by a human can be copyrighted under United States law, which excludes pictures and artwork created by animals or by machines without human intervention”
and
“because copyright law is limited to ‘original intellectual conceptions of the author,’ the copyright office will refuse to register a claim if it determines that a human being did not create the work”.
In a separate matter, People for the Ethical Treatment of Animals (PETA), an American animal rights non-profit organization, sued Mr Slater in September 2015 before the California District Court acting as a ‘next friend’ for Naruto. This action followed Mr Slater’s publication of the photograph in a book, “Wildlife Personalities”, published by Blurb Inc. Blurb was also a defendant in the action.
PETA requested that the District Court (i) find Naruto as the author of the photograph and therefore owner of the copyright, and (ii) grant the administration and management rights in the photograph to PETA, following which PETA would use the profits from the commercialisation of the photograph for the benefit of Naruto and other macaques in Indonesia.
At first instance before the California District Court in January 2016, it was held that a monkey (as a non-human) could not be considered an ‘author’ within the meaning of the US Copyright Law. The decision was appealed to the Ninth Circuit Court which, in 2017, affirmed the decision of the District Court and made clear that animals, as non-humans, could not be considered ‘authors’.
7. A Piece of American Cheese
The position that a non-human, such an as animal, cannot be considered an ‘author’ of a work for the purposes of copyright principles is well-settled across jurisdictions.
However, jurisdictions are also considering the same question as the Sarony-Burrow-Giles case 141 years ago but within the context of AI rather than cameras: where AI technology is used by a human to create a work, does copyright subsist in that work?
On 5 September 2023, the Review Board of the United States Copyright Office issued a final decision refusing the application for registration Jason Allen of “Théâtre D’opéra Spatial” (the Work) as a copyright work:
Source: https://www.copyright.gov/rulings-filings/review-board/docs/Theatre-Dopera-Spatial.pdf
From a general public perspective, this image is famous for having won the digital art division of the 2022 Colorado State Fair fine art competition when it was not known by the judges that the Work had been created by AI.
From copyright perspective, it was the ‘human touch’ threshold that is particularly important since the Work was created through a combination of prompts into Midjourney (a generative artificial intelligence program) Mr Allen and then edits and refinements by Mr Allen through Adobe Photoshop and Gixapixel AI.
In reviewing how the resulting Work was created, the Review Board considered how Midjourney operated and what steps Mr Allen undertook in the creation of the Work. It was stated:
“It is the Office’s understanding that, because Midjourney does not treat text prompts as direct instructions, users may need to attempt hundreds of iterations before landing upon an image they find satisfactory. This appears to be the case for Mr. Allen, who experimented with over 600 prompts before he “select[ed] and crop[ped] out one ‘acceptable’ panel out of four potential images … (after hundreds were previously generated).”
The Review Board also considered in detail what input was provided by Mr Allen to the original Midjourney Image (Below Top) to result in the Work submitted for copyright registration (Below Bottom):
Midjourney Image
The Work submitted for registration
Source: https://www.copyright.gov/rulings-filings/review-board/docs/Theatre-Dopera-Spatial.pdf
The Review Board found that the Work contained more than a de minimis amount of content generated by AI and stated:
“when an AI technology receives solely a prompt from a human and produces complex written, visual, or musical works in response, the ‘traditional elements of authorship’ are determined and executed by the technology - not the human user.”
The Review Board noted that since Mr Allen was unwilling to disclaim the AI-generated elements of the Work, it could not be registered as a copyright.
This line of reasoning, which considered a distinction between human and non-human generated elements in a work was in line with the decision of the District of Columbia District Court on 18 August 2023 in Thaler v. Perlmutter.
Dr. Stephen Thaler is the founder of Imagination Engines Incorporated (an artificial neural network technology company) and developer of various AI systems, including one named the “Creativity Machine”. This was used to generate a work titled “A Recent Entrance to Paradise”:
Source: Thaler v. Perlmutter, No. 22-CV-384-1564-BAH (United States District Court for the District of Columbia – 18 August 2023)
When Dr Thaler applied to register the image with the United States Copyright Office, listing the AI as the author, the application was denied on the grounds that copyright protection requires human authorship and that this work had been autonomously created entirely by AI.
Dr. Thaler appealed the decision to United States District Court for the District of Columbia. The court unanimously affirmed the Copyright Office’s decision and stated that:
“Copyright is designed to adapt with the times. Underlying that adaptability, however, has been a consistent understanding that human creativity is the sine qua non at the core of copyrightability, even as that human creativity is channeled through new tools or into new media”.
The court also relied on the United States Copyright Act’s stipulation that protection is granted only to “original works of authorship”, interpreting the term “author” as historically and judicially confined to human beings.
This application of a ‘human input’ threshold also resulted in the initial refusal by the United States Copyright Office of a copyright application filed on 5 August 2024 for a work titled “A Single Piece of American Cheese”:
The application was filed by Kent Keirsey, and the work was created using Invoke AI, an AI platform founded by Mr Keirsey.
Following the refusal, an application for reconsideration was filed, in support of which materials and a video were submitted which showed the process and steps in the creation of the final work from the original image which had been generated by Invoke in response to the prompt entered into the AI:
Original Invoke AI Image
The Work submitted for registration
It was submitted by Mr Keir that, as part of this process, he “selected, coordinated, and arranged numerous AI generated image fragments into a single, unified image” and also relied on ‘inpainting’ which is described as follows:
“Inpainting in generative AI is the process of selectively modifying or regenerating parts of an image while maintaining consistency with the surrounding elements. Invoke’s canvas is uniquely designed to give artists a high degree of control, allowing them to precisely define where and how regions are regenerated.”
Having reconsidered the application, the United States Copyright Office granted registration to “A Single Piece of American Cheese” on 30 January 2025 but excluding the AI generated elements. The basis of the claim, as recorded on the copyright registration, is “Selection, coordination, and arrangement of material generated by artificial intelligence”.
In a public guidance “Copyright and Artificial Intelligence – January 2025” issued on 29 January 2025, the United States Copyright Office stated:
“For a work created using AI, like those created without it, a determination of copyrightability requires fact-specific consideration of the work and the circumstances of its creation. Where AI merely assists an author in the creative process, its use does not change the copyrightability of the output. At the other extreme, if content is entirely generated by AI, it cannot be protected by copyright. Between these boundaries, various forms and combinations of human contributions can be involved in producing AI outputs”.
8. Conclusion
It would appear, for the United States and various other countries, a doctrine is emerging that copyright can subsist in a work created with AI provided that:
- there is a sufficient level of contribution by a human; and
- the nature of contribution is to be considered on the facts of each case.
What is likely to emerge is a continuing requirement to have a disclaimer of the AI-generated element of the particular work, whether an image or other form of work in which copyright can subsist.
For instance, if an individual merely instructs an AI tool to generate a comprehensive research paper by providing only the title, that person cannot be considered the author. However, if the individual uses creative efforts by providing detailed instructions, guiding the AI with innovative input, overseeing and refining the final output, then authorship may be attributed proportionately to the extent of the human contribution.
This is a rapidly developing area of law and, as has been seen with other areas of law, jurisdictional differences and overlaps will need to be considered when advising in this area.
As at the time of writing, the principles suggested by the United States Copyright Office in its 2025 position paper are very useful guidelines:
- The use of AI tools to assist rather than stand in for human creativity does not affect the availability of copyright protection for the output.
- Copyright protects the original expression in a work created by a human author, even if the work also includes AI-generated material.
- Copyright does not extend to purely AI-generated material, or material where there is insufficient human control over the expressive elements.
- Whether human contributions to AI-generated outputs are sufficient to constitute authorship must be analysed on a case-by-case basis.
- Based on the functioning of current generally available technology, prompts do not alone provide sufficient control.
- Human authors are entitled to copyright in their works of authorship that are perceptible in AI-generated outputs, as well as the creative selection, coordination, or arrangement of material in the outputs, or creative modifications of the outputs.
With these principles in mind, it is recommended that where AI is relied upon to generate a particular work, a careful and detailed record should be kept of the human contribution into the final form of the work in order to:
- identify the degree of human contribution; and
- consider whether this would be sufficient to result in a copyright in the work under the prevailing law and practice in which copyright in the work is sought to be registered, or enforced against others.
On this last point, the increasing speed with which AI is taking a key role in many aspects of our public and private lives is such that disputes on the extent to which copyright in ‘human authored AI generated’ works can be enforced against others will become increasingly common.
In the meantime, for a work wholly created by AI then the position remains as clear as Naruto’s grin – for a work created wholly by a non-human, copyright will not subsist.
If you would like further information on this topic, please contact James Dunne (j.dunne@hadefpartners.com), Rami Omar (r.omar@hadefpartners.com) IP@hadefpartners.com, or your usual Hadef & Partners contact.