ABSTRACT:

The recent Sony World Photography Award sparked a heated debate among photographers and artists worldwide. Boris Eldagsen’s winning entry in the creative photography category was met with controversy after he revealed that it was created using an artificial intelligence image generator. In this article, I will explore how competitions can adapt to the evolving digital art landscape shaped by AI (Artificial Intelligence). What factors must be considered? Is establishing distinct criteria for AI necessary? Should AI be viewed solely as a creative tool or as a work of art? The importance of addressing these issues cannot be overstated. During my tenure as the chair of the panel of judges for the Vision Petron National Student Art Competition’s Digital Art Category in March of this year, we faced certain nuanced issues. The fact that AI is now involved in the creation process created apprehension among the panel. In this paper, I aim to delve into crucial discussions in this field and their impact on photography and digital art competitions. It is important to recognize the unique characteristics that AI brings to the creative process and consider how these should be evaluated. As AI becomes more prevalent in the creation of visual art, it is likely that we will see new forms emerge. Competitions must be prepared to adapt to these changes and embrace innovation in the field. 

Keywords: promptography, digital art competitions, controversy, AI-generated image, Sony World Photography Award, AI image generator, photorealistic imagery 

Controversy Erupts Over AI-Generated Image Winning Prestigious Photography Award 

On April 13 in London, Berlin-based German photographer Boris Eldagsen was awarded the prestigious Creative Photography category of the 2023 Open Competition of the Sony World Photography Award. His winning entry, titled Pseudomnesia: The Electrician, was created using DALL-E 2, an AI image generator developed by OpenAI. 

Figure 1 

Boris Eldagsen’s Pseudomnesia: The Electrician 

Eldagsen’s aim was to challenge the contest and spark discussion about a technology that has the potential to alter our perception of photorealistic imagery. After the announcement, he declined to accept the award, leading to a strained relationship between him and the organizer. Despite not being invited, Eldagsen took the stage during the ceremony to address the audience. 

He said AI images are not photographs and therefore should not be considered in competitions designed for camera-based practitioners. On his website, Boris stated that the judges lacked the ability to differentiate between photographic images and those generated by AI machines. He argued that his creations are not photographs but “images” artificially synthesized with photographic elements as visual language (2023). 

The disclosure raised multiple inquiries regarding the characteristics of artwork produced by AI and raised doubt about the definition of photography, whether it should be distinguished from its medium. The photographer expressed disappointment stating, “photographic language has dissociated itself from its medium, which is disheartening for photography because AI now defines the future of the field” (Fotopodden, 2023). 

According to World Photography Award officials, they are aware that the image is AI generated. The Creative Photography category in its tradition has always been open for experimentations. A spokesperson for the Sony World Photography Awards clarified: 

 

During our various exchanges with Boris Eldagsen ahead of announcing him as the Creative category winner in the Open competition on 14 March, he had confirmed the ‘co-creation’ of this image using AI. In our correspondence he explained how following ‘two decades of photography, my artistic focus has shifted more to exploring creative possibilities of AI generators’ and further emphasizing the image heavily relies on his ‘wealth of photographic knowledge’. As per the rules of the competition, the photographers provide the warranties of their entry (artforum, 2023). 

The officials of the Award have announced the suspension of all activities with him, citing his deliberate attempts to mislead them, which has rendered the warranties he provided invalid. Eldagsen, on the other hand, contests this assertion and maintains that he participated as a cheeky monkey. He stated: I opted to submit an image that was artificially created. I saw myself as a hacker attempting to uncover weaknesses in the system. My intention was not to exploit the competition, but rather to highlight an area in need of attention (Eldagsen, 2023). 

The controversy surrounding Eldagsen’s entry has sparked an intense debate within the photography community. Some argue that AI-generated images should not be allowed in photography competitions, while others believe that the use of technology is essential for the evolution of the art form. 

In the advent of photography, artists had similar apprehensions like today with AI, in his website Derek Murphy noted: “Baudelaire called photography the refuge of failed painters with too little talent. Van Gogh said photographs could never capture that human spark, so he started focusing more on painting portraits” (Murphy, 2022). 

Eldagsen’s actions have also raised questions about the responsibility of artists and creators to disclose the use of AI in their work. While some feel that it is important to be transparent about the tools and techniques used, others argue that the focus should be on the outcome and the impact it has on the viewer. While John Lennox puts it in his book titled 2084: “The real problem with AI, then, is . . . the likelihood of our blindly depending on machines, lulled to trust them by bad metaphors. The danger is that computers will fail us and do so in bad ways”(2020). 

Regardless of where one stands on the issue, the use of AI in art and photography will continue to be a hotly debated topic in the years to come. As technology advances and artists push the boundaries of what is possible, it will be the responsibility of organizations such as the Sony World Photography Awards to navigate these uncharted territories and establish definitive criteria for participation in their competitions. 

Navigating the Intersection of Photography and AI in Competitions 

One of the most significant concerns surrounding the use of AI in photography competitions is the ethical implications. AI can be used to manipulate images and create something that was not initially there, which raises questions about the authenticity and truthfulness of the images submitted to competitions. While AI can enhance an image, it can also be used to deceive viewers, that may lead to a loss of trust in the authenticity of the photography industry. 

Boris asserted that photography and AI should not be lumped together. The next step would be to talk about the correlation between promptography and photography, and whether they should be grouped together in a single museum, festival, gallery, or competition. This matter is intricate, and I cannot provide a definite response. However, I can state that the simplistic views held by those who advocate for the return to traditional techniques and those who claim that promptography is equivalent to photography are misguided. To reach a satisfactory conclusion, we need to delve deeper into the issue (Parshall, 2023). 

His concern was valid, as he predicted that thousands of AI-generated images could flood photography competitions in a year’s time. Boris expressed: Initially, I was aware of a single software that could create images based on textual cues, but I am astounded by the rapid expansion of internet resources in the last year. There was a significant surge, and now the progress is continuously picking up momentum (Eldagsen, 2023). 

This raises a challenge in distinguishing between photographs, which are captured using light and lenses, and AI images, which are created through prompt inputs. The disparity between mediums is what reveals the most significant differences. However, as judges in a competition, we only have access to the final output and cannot see the process behind its creation. This leaves us questioning how we can accurately differentiate. As time passes, deciphering these differences will only become more challenging than it is today, as technology progresses exponentially over time. 

Eldagsen used the term ‘promptography’ to refer to the image, pointing out how platforms like DALL-E 2 and ChatGPT utilize “prompts” or precise user instructions to generate personalized visuals or content. According to Eldagsen in an interview with the BBC, while promptography is done with prompts, photography is done with light. He believes it is crucial to distinguish between the two and have an open discussion within the photography community. The question remains whether the realm of photography is inclusive enough to encompass this type of imagery, given that the visual language is similar (Seymour,2023). 

Reflections on the Future of Competitions 

Creating digital art, like photography, can be a complex and intricate process. During the judging phase of the Vision Petron Digital Art competition in March, our primary focus was to identify the entries that effectively conveyed a storyline and communicated the theme. As we reviewed the shortlisted entries, we realized that we needed access to the layered files to understand how each artwork was created. Our goal was to locate areas where AI technology was utilized and explore how it was employed in the artwork. 

Looking for the AI footprints: for some, it is easy to say that they can recognize which image is created using AI, at least for now. As technology becomes more sophisticated, the difference between analog and AI may become permanently undetectable. 

What properties and qualities are we looking for and how do we deconstruct the new form that this digital art has evolved into? One way of looking at it is to utilize the classic Aristotle’s four causes, namely, material cause, formal cause, efficient cause, and final cause. The material cause (what something, say a table, is made of); the formal cause (the form that determines the shape of the table); the efficient cause (what brings the table into being–a carpenter); and a final cause (the purpose for which the table was made–to use for dining) (Falcon, 2023). 

Let us explore how this concept could be relevant to the field of digital art and artificial intelligence. 

1. Material cause: This refers to the material or substance from which an object is made. For digital art, the material cause can be the software or hardware used to create the artwork. 

2. Formal cause: This refers to the form or shape that an object takes. For digital art, the formal cause can be the style or technique used to create the artwork. 

3. Efficient cause: This refers to the process or action that leads to the creation of the object. For digital art, the efficient cause can be the use of AI technology to assist in the creative process. 

4. Final cause: This refers to the purpose or end goal of the object. For digital art, the final cause can be to convey a message or tell a story through visual means. 

I want to emphasize the significance of the final cause, as it gives an artwork its purpose and meaning. Without context and intent from the artist, such creations are merely arbitrary.   

To further examine the legitimacy of considering AI as an art form, we can utilize the methodology of sufficient and necessary conditions. The necessary condition requires the use of generative art software in the creation process, while the sufficient condition mandates that an artist creates the AI image with the intention of conveying their unique vision. For an AI image to be recognized as true art, it must fulfill both the necessary and sufficient conditions.  

As Harrison and Wood stated: It may be beneficial to examine discourses not just based on their expressive worth or formal changes, but by their modes of existence. The ways in which discourses circulate, gain value, are attributed, and are adopted differ among cultures and are adjusted within them. The means in which they are expressed through social relationships can be better comprehended through examining the author-function and its alterations than through the topics or ideas that discourses generate (Harrison & Wood, 2007). 

However, Bernard Marr (2023) argues that the line between humans and machines is becoming increasingly unclear. Marr claims that creativity has always been a distinguishing factor between the two, but the emergence of a new type of artificial intelligence called “generative AI” is challenging this notion.  

It is crucial to acknowledge the author’s role and the means through which an object is articulated or produced, hence the significance of comprehending the agency and mechanism as a method to interpreting an object, in this case digital art, and the necessity of identifying their connection. 

For digital art competitions, we can expect AI to be involved in a great portion, if not all, of the process. An acceptable amount of AI involvement should be part of the discussion. Assuming participants use AI, I suggest technical proficiency in other design and drawing software be given more consideration. As most digital art contests in the Philippines target art and design students, judges would like to see the mastery that the participants have developed from their school training. Prompt engineering may also be part of the judging together with the other criteria. 

For photography competitions— to better inform judges in selecting winners; I propose that shortlisted artists provide a comprehensive documentation of their process; including the raw file and every step from pre-production to post-production— if ever post-production is permitted. Post-production can take on various forms depending on the software used by the artist. One such tool is the AI image generator, Midjourney, which allows the artist to blend their own images with elements produced by AI. This documentation should be mandatory for creative photography categories. If AI-generated elements are being considered in the process for the creative photography category, which in SPWA’s statement that they did allow, clear guidelines must be provided. 

One potential solution to the challenge of distinguishing between photography and promptography is to create a new category specifically for AI-generated images. This would allow for a clear differentiation between the two mediums and avoid any confusion in the judging process. Additionally, this could provide a platform for promptography artists to display their work and gain recognition. However, there must be clear guidelines and criteria for this new category to ensure fairness and consistency in judging. Another option could be to integrate promptography into a broader digital art category, which may already exist in some competitions. This would acknowledge the unique process of promptography. It is crucial for the photography community to engage in open and ongoing discussions about the relationship between photography and AI to ensure a clear understanding and progression in the field. 

In the end, the use of AI in photography competitions should be approached with caution and promoted with transparency. Competitions should have clear guidelines for the use of AI, and participants should disclose any use of AI in their images. Judges and organizers should also be well-versed in AI technology to ensure that ethical concerns are addressed and that the creative process is not stifled. 

In conclusion, the landscape of digital art competitions is evolving rapidly with the integration of AI technology. While it may be difficult to distinguish between traditionally produced digital art and entirely AI-produced artwork presently, the incorporation of AI technology into the creative process of digital art may become a distinct quality of the art form. As digital art continues to develop, it is essential to identify the new properties and qualities that it develops into. The final cause of digital art, that is, the purpose and context provided by the artists, must be given consideration in digital art competitions. As technical proficiency of traditional art-making processes still holds significant value, judges must weigh in on the appropriate level of AI involvement. The future of digital art competitions may see the integration of AI technology in varying degrees, but it is the artists’ mastery and expression that will always be the key factor in conveying a message or storytelling. 

References 

Artforum. (2023, April 19). Sony World Photography Awards Winner Reveals Entry Was AI- 

Generated, Rejects Prize. Artforum. [https://www.artforum.com/news/sony-world-photography-award-winner-reveals-entry-was-ai-generated-rejects-prize-90416] (https://www.artforum.com/news/sony-world-photography-award-winner-reveals-entry-was-ai-generated-rejects-prize-90416) 

Eldagsen, B. (2023, April 26). I used AI in a competition for photographers. It won. Newsweek.  

[https://www.newsweek.com/ai-photography-contest-sony-art-1796455] (https://www.newsweek.com/ai-photography-contest-sony-art-1796455) 

Falcon, A. (2023). Aristotle on Causality. In E. N. Zalta & U. Nodelman (Eds.), The Stanford  

Encyclopedia of Philosophy (Spring 2023 Edition). [https://plato.stanford.edu/archives/spr2023/entries/aristotle-causality/] (https://plato.stanford.edu/archives/spr2023/entries/aristotle-causality/) 

Fotopodden. (2023, March 23). AI defines the future of photography. [Video]. YouTube.  

[https://www.youtube.com/watch?v=KbLAcsXGXnk](https://www.youtube.com/watch?v=KbLAcsXGXnk) 

Harrison, C., & Wood, P. (2007). Art in theory, 1900-2000: An anthology of changing ideas.  

Blackwell. 

Lennox, J. (2020). 2084: Artificial intelligence and the future of humanity. Zondervan. 

Marr, B. (2023, March 27). The intersection of AI and human creativity: Can machines really be  

creative? Forbes. [https://www.forbes.com/sites/bernardmarr/2023/03/27/the-intersection-of-ai-and-human-creativity-can-machines-really-be-creative/?sh=6df037d03dbc] (https://www.forbes.com/sites/bernardmarr/2023/03/27/the-intersection-of-ai-and-human-creativity-can-machines-really-be-creative/?sh=6df037d03dbc) 

Murphy, D. (2022, December 6). The creative war against AI art has already been lost.  

Creativindie. [https://www.creativindie.com/the-creative-war-against-ai-art-has-already-been-lost/] (https://www.creativindie.com/the-creative-war-against-ai-art-has-already-been-lost/) 

Parshall, P. (2023, March 26). How this AI image won a major photography competition.  

Scientific American. [https://www.scientificamerican.com/article/how-my-ai-image-won-a-major-photography-competition/] (https://www.scientificamerican.com/article/how-my-ai-image-won-a-major-photography-competition/) 

Seymour, T. (2023, April 18). The camera never lies. Creator of AI image rejects prestigious  

photo award. The Art Newspaper. [https://www.theartnewspaper.com/2023/04/18/the-camera-never-lies-creator-of-ai-image-rejects-prestigious-photo-award] (https://www.theartnewspaper.com/2023/04/18/the-camera-never-lies-creator-of-ai-image-rejects-prestigious-photo-award) 

ABOUT THE AUTHOR: 

Robert Nelson Besana is an artist, musician, and educator. He is the Executive Director of the School of Multimedia and Arts at NU-Asia Pacific College. He is also a member of the Commission on Higher Education (CHED) Technical Committee for Multimedia Arts. Since 2008 he has held eleven solo exhibitions in both local and international galleries and museums. He holds an MA in Fine Arts and Design from the Philippine Women’s University (Corresponding author: robertb@apc.edu.ph).