The Audience of One: Readers Turn to AI for Custom Fiction
While the publishing industry fights AI-generated books, a grassroots movement of readers is using chatbots to build hyper-personalized narratives.
A growing number of readers are bypassing traditional publishing to create personalized, interactive fiction using AI chatbots. This shift toward reader-driven generation suggests a decoupling of literary value from entertainment value, as users prioritize tailored experiences over professional authorship.
Data from the WildChat dataset reveals the scale of this trend. An analysis of over 500,000 ChatGPT prompts showed that more than a third of interactions involve fiction generation, spanning prose, poetry, fan fiction, erotica, and role-play. Researchers estimate that approximately 7 percent of all chatbot users engage in fiction generation, a figure that remains significant even when controlling for the disproportionate activity of 'power users.' However, the definition of 'fiction' varies by study; a 2025 report by OpenAI and the National Bureau of Economic Research suggests only 1.4 percent of prompts are fiction-related, though that specific figure excludes role-play.
The Professional Divide
This grassroots adoption is unfolding against a backdrop of intense volatility in the professional publishing world. While readers embrace the technology, the industry has reacted with hostility, with some publishers canceling multi-million dollar deals and withdrawing novels due to suspected AI use.
For professional authors and publishers, AI is often viewed as a threat to the integrity of the craft. In contrast, the community of AI fiction users views the technology's flaws as assets. Software engineer and AI fiction user Petra Ferraz de Novaes notes that the unpredictability of the models is a primary draw. "People say that AI hallucinates a lot," Novaes said. "But in this case, it’s probably what you want."
A Shift in Consumption
This movement represents a fundamental change in how stories are consumed. Rather than reading a static text designed for a mass audience, users are leveraging commercial models and open-source tools like SillyTavern to build endless, unpredictable narratives tailored to their specific desires.
According to Melanie Walsh, an assistant professor at the University of Washington Information School, the lack of a human author is not a deterrent for these users. "They know [the stories are] produced by an AI model, and they don’t care," Walsh said. "In fact, that seems to be a big part of the attraction."
The Future of Narrative
As these tools become more sophisticated, the industry may face a future where a significant portion of fiction is generated on-demand for a single audience of one. This suggests a transition from the era of the 'bestseller' to an era of hyper-personalized consumption, where the value of a story lies in its immediate relevance to the reader rather than its standing in the literary canon. What remains to be seen is how traditional publishing will adapt—or if it will simply coexist with a parallel, invisible industry of AI-generated private libraries.