In an innovative fusion of technology and literature, the first-ever AI-powered literary award competition is set to launch this March, bringing to life the vision I outlined in my article “The Future of Literary Award Competitions: Why AI Should Replace Human Judges,” published in Arab World Books on 6 November 2024. Led by Timu Bozsolik-Torres, a software engineer, data scientist, and former Google innovator behind Gemini—who supported the arguments set forth in my article—in collaboration with MyPoolitzer.com, this initiative tackles the subjectivity and bias of traditional judging. It offers a fairer, more inclusive platform for authors, including the many forgotten voices striving for recognition alongside established names.
AI as a Sports Official
While this will be the first AI-powered writing competition of its kind, artificial intelligence already has a presence in various facets of competitive evaluation, including the world of sports, where it is increasingly being explored as a tool for officiating. After a long period of human dominance in judging, AI is now being tested as a powerful tool for assessment, offering greater precision, consistency, and impartiality.
One of the most recent examples of this technological advancement occurred during the highly anticipated Oleksandr Usyk vs. Tyson Fury rematch for the unified heavyweight world championship. As reported by DAZN, an experimental AI scoring system was introduced alongside the three traditional judges. While the official human judges scored the fight 116-112 in favour of Usyk, the AI system saw the defending champion winning by a slightly wider margin of 118-112. The AI awarded Fury only four rounds—the first, second, fifth, and twelfth—suggesting a different interpretation of the fight’s pacing and dominance.
The introduction of AI in sports judging represents a significant leap forward in enhancing fairness and minimizing human error. As technology continues to evolve, AI-powered officiating is poised to become a game-changer in competitive sports, offering athletes, coaches, and fans a more precise and data-driven evaluation system.
AI as an Advice Expert
AI is also revolutionizing the legal system. According to the BBC, “increasingly, law firms are turning to AI to help them wade through vast amounts of legal data.” Additionally, researchers have claimed that “an artificial intelligence system has correctly predicted the outcomes of hundreds of cases heard at the European Court of Human Rights.”
Moreover, the integration of AI in legal advice is on the rise, with clinics like the Westway Trust’s Cost of Living Crisis Clinic in London adopting this technology to assist clients who may struggle to afford legal representation against wealthier opponents.
Adam Samji, a paralegal adviser featured in the BBC report, highlights this technology’s efficiency: “We spend a couple of minutes going through [the documents] and redacting the client’s personal information. We then upload it onto an AI model, which provides us with the necessary insights. It usually returns the results in about 10 to 15 minutes.” He adds, “It saves us hours of manual work. As paralegal volunteers, we can use our time more efficiently to better serve our clients.”
This demonstrates how technology is playing an increasingly vital role in making legal support faster, fairer, and more impactful.
AI: a Fairer Judge in Writing Competitions
AI also has the potential to reshape literary award competitions, which have long been fraught with disputes.
Such controversy is inevitable, as human judges are naturally influenced by subjective factors such as personal tastes, cultural backgrounds, national sentiments, and gender biases. Literary award results often reflect the preferences of a small panel rather than the broader literary community. If the same books were submitted to different judging panels, the outcomes would likely vary significantly—highlighting the relativity, inconsistency, and unpredictability of human evaluation.
Beyond subjectivity, human judges are also limited in their ability to read large volumes of material. As noted by Thuy On, a seasoned literary judge: “Most judges read several chapters to gauge the writing quality. If the prose or poetry piques their interest, they read on. If not, they move on to the next.”
This approach risks overlooking works that require more time to be fully appreciated. Unlike humans, AI judges can process entire manuscripts, ensuring that every submission is read in full and eliminating premature dismissals. Additionally, AI allows for the inclusion of all published books rather than just a select few among tens of thousands—preventing literary awards from being decided based on a mere fraction of the available literature.
A New Opportunity for Forgotten Writers
For all these reasons, I advocated for the use of artificial intelligence instead of human judges in writing competitions in my article. This proposal is now becoming a reality through the first AI-powered competition, which combines both robot and human judges. This groundbreaking event presents a unique opportunity for unpublished authors to gain recognition and potentially launch their writing careers.
Authors worldwide will have the opportunity to submit their original, unpublished works of fiction in any genre, written in English and no longer than 80,000 words. The submission window is open until March 31, with the winner announced by the end of April.
The submission process is simple. Authors will first submit a synopsis and the first 20 pages of their manuscript. A shortlist of authors will then be invited to submit their full manuscripts for final evaluation, which will be assessed by both an AI system and a panel of human judges from the publishing industry. This hybrid approach ensures a balanced and thorough evaluation process, combining the efficiency of AI with the nuanced understanding of human experts.
Balancing Literary Excellence and Sellability
The new AI-powered literary award competition is a promising initiative. However, as this innovative competition evolves, addressing a few concerns could enhance its fairness and effectiveness.
One such concern is the restriction of participating manuscripts to just 20 pages. This approach can lead to premature rejection. Given AI's unparalleled data-processing capabilities, it should be leveraged to evaluate entire manuscripts, ensuring fairer and more reliable results.
Another issue is the 80,000-word limit for submissions, which can lead to the exclusion of many works—including my unpublished debut English novel, which exceeds 126,000 words. Many renowned novels surpass this limit; for example, Harry Potter and the Prisoner of Azkaban has 107,253 words, while Harry Potter and the Goblet of Fire reaches 190,637 words. I encourage the organizers to adopt a more flexible word count policy to accommodate a wider range of submissions.
A further concern is the final judging process, which involves both AI and human judges. I personally advocate for AI to handle the entire evaluation, with human judges intervening only when necessary. If AI can assess thousands of submissions, there is no reason it should not be capable of evaluating the few shortlisted works. Unless AI determines the winner, the results will always be open to dispute.
Another issue is the competition’s emphasis on the "sellability" of manuscripts. According to the organizers' website: "Iris will assess the sellability of your manuscript, compare it to others in the genre across a variety of metrics, and identify opportunities for improvement."
While commercial viability is important, overemphasizing sellability risks favouring established authors with proven sales records while sidelining emerging voices. The evaluation process should prioritize depth of ideas and a manuscript’s ability to resonate with readers on a human level.
A related concern is the competition’s preference for bestseller styles, as highlighted on the organizers’ website: "Iris evaluates how closely your work aligns with your specified genre and how well your style resembles that of a typical bestseller across 30+ dimensions."
This raises the possibility that participants could manipulate the AI by mimicking commercially successful styles, potentially winning at the expense of writers who strive to develop unique voices and produce innovative work.
