Can AI Really Write a Novel? Assessing the Future of Automated Storytelling
Rifki Aria
Table of Contents
- How AI Novel Writing is Revolutionizing Literature
- The Role of AI in Automated Storytelling
- Is AI Ready to Take Over?
- References
How AI Novel Writing is Revolutionizing Literature
AI novel writing has sparked a fascinating debate about the future of literature and creativity. Who would have thought that novels written by AI could win a national writing competition in China and a prestigious award in Japan? The utilization of AI in novel writing has become more prevalent over time. Authors have found AI-generated prose and paragraphs to be much less weird and corny than they used to be not so long ago. As a result, they have grown dependent on AI-generated literature generators.
Intriguingly, Reuters reported that over 200 books available on Amazon’s e-book platform listed ChatGPT as either their author or co-author as of February 2023, implying the ever-increasing prevalence of AI in creative writing practices. Such a trend has threatened human authors. They can no longer rest and expect their ideas to remain exclusive to themselves forever; others might easily use those ideas and even turn them into real products instantly. It is all about finding the right prompts to create best-selling novels.
Has AI been ready to replace human novelists? Can AI truly write a novel that resonates with readers? Let’s find out as we take you to explore AI utilization in novel writing.
The Role of AI in Automated Storytelling
AI’s capacity to generate content is a potential game-changer for the creative writing industry. Human authors have begun exploring the technology’s automated storytelling capacity to overcome writer’s block. They only need to craft effective prompts that may grant them quality responses, which are phrases or paragraphs. To some extent, those responses can even be as long as a novel draft.
The quality of AI-generated novels depends on several features of AI. One key feature that makes AI suitable for writing novels is its natural language processing capacity. It enables AI machines to translate and produce wording and paragraphs that are natural to human audiences. However, they require sufficient input of words and phrases to be effective. This is because, essentially, AI language models such as GPT-3 are mere word-prediction machines with no adequate capacity to produce human-natural languages.
A pseudonymous researcher and writer, Gwern Branwen, claimed that platforms for AI writing novels, such as Sudowrite, need the user’s sympathy to function adequately, expecting humans’ attempts to understand the machine’s potential capacity for creative writing. This finding complements the promulgation that AI-generated literature lacks cultural diversity and the capacity to execute out-of-context ideas. Nonetheless, these worries do not nullify the belief in the potential of AI in creative writing.
When Rie Kudan won the 170th Akutagawa Prize, a prestigious award for emerging authors in Japan, little did the public know that AI wrote 5% of her award-winning novel, Tokyo-to-Dojo-to. However, she admitted that the parts written by AI did not disrupt the natural flow of her story, which she wrote herself, thus confirming that AI writing novels might still not have what it takes to be fully responsible for producing qualified novels.
Nevertheless, Kudan’s admission prompted judges of the award to consider changing submission requirements for future contests because contestants might feel cheated by the use of AI in creative writing activities, further reflecting the worries of human authors for AI writing novels.
A similar case occurred in China when Professor Shen Yang of Tsinghua University utilized ChatGPT to produce a nearly 6,000-character Chinese-language novel, Land of Memories, which eventually won the second prize in the Jiangsu Youth Popular Science Fiction Competition. Shen crafted 66 prompts to gather approximately 4,300 characters before rearranging those characters for the final draft.
Interestingly, one of the judges noticed that Shen employed AI in the novel, as he complained that the work lacked emotion, thus preventing him from favoring the novel. This event confirmed that AI-generated literature may not be as emotionally engaging as the works produced by human writers.
Is AI Ready to Take Over?
After reviewing the two case studies above, it is safe to assume that novel writing AI still lacks two things to be considered capable competitors to human authors: creativity and emotional resonance. As suggested earlier, AI machines may be incapable of exploring unorthodox topics, such as religion or race-sensitive ones. The ethical standards input into these machines may prevent them from working on such topics, thus making them unreliable to authors whose niches are within the territory of controversial topics.
In terms of emotional resonance, it seems that human authors cannot expect their AI-generated literature to come up with works that shake up their audiences’ emotions. Consequently, it is rather challenging to determine if the audience can resonate with such works. This particular worry syncs up with the common perception that AI lacks the capacity to understand people’s emotions.
That said, it is best not to overhype AI’s creative writing capacity. It is true that AI can become a highly potential writing assistant due to its ability to retrieve, analyze, and present big data, which can be helpful in creative writing activities. However, it is safe to say that AI alone cannot produce proper literature without human supervision. While it may have the ability to generate texts that appear to be human-written, those texts may not be as in-depth or complex as the ones written by real humans.
In conclusion, we do not think that AI is ready to replace human authors, although it might not be impossible in the future. However, it is understood that writers should begin adopting AI in their creative writing processes due to the potential benefits the technology may offer, including the opportunity to overcome writer’s block. To avoid missing any updates on this matter, stay updated and join our discussion on how AI may impact creative writing.
References
Abeer (2024, February 14). Will AI write books in the near future? GetGenie. Retrieved July 7, 2024, from https://getgenie.ai/will-ai-write-books-in-the-near-future/
Ali, M. (2023, December 1). Would I use AI to write my novels? I’d get better results from a monkey with an iPhone. The Guardian. https://www.theguardian.com/commentisfree/2023/dec/01/would-i-use-ai-to-write-my-novels-id-get-better-results-from-a-monkey-with-an-iphone
Chik, H. (2023, December 20). A Chinese professor used AI to write a science fiction novel. Then it was a winner in a national competition. MSN. https://www.msn.com/en-xl/news/other/a-chinese-professor-used-ai-to-write-a-science-fiction-novel-then-it-won-a-national-award/ar-AA1lMwUa
Dzieza, J. (2022, July 20). The great fiction of AI. The Verge. https://www.theverge.com/c/23194235/ai-fiction-writing-amazon-kindle-sudowrite-jasper
Robertson, A. (2023, May 24). I tried the AI novel-writing tool everyone hates, and it’s better than I expected. The Verge. https://www.theverge.com/2023/5/24/23732252/sudowrite-story-engine-ai-generated-cyberpunk-novella
Tipu, A. (2024, May 1). Will artificial intelligence replace my job? How to thrive in an AI era. The Inclusive AI. https://theinclusiveai.com/how-to-thrive-in-an-ai-era/.
Yao, D. (2024, February 6). Author admits to using AI to write award-winning novel. AI Business. https://aibusiness.com/nlp/author-reveals-using-chatgpt-for-award-winning-novel#close-modal
