Shy Girl is the first book to have been dropped by a major publishing company following such accusations of undisclosed AI use, one of the main reasons why the discussion has grown into such an internet minefield, and it begs some very important questions regarding the future of the publishing industry. Namely, how did the book make it to the publishing stage without the reportedly blatant use of AI being flagged? What even is the objective stance on AI use within the publishing industry? And how do publishing houses intend on upholding their commitment to championing original works in a society where AI is becoming both more widely spread and is wildly unregulated?
For those who don’t know about the Shy Girl controversy, let me catch you up. The novel, a female-rage horror narrative about a woman held hostage by a man she meets online who keeps her as his pet, was originally self-published by Ballard in February 2025, before it was acquired by UK publishing house Hachette, who published the novel in November. It was scheduled to be published this year in the US but suspicions began to be raised online by readers in YouTube reviews and TikTok and Reddit posts about whether generative AI had been used to write parts of the novel. This was put down to the novel’s reportedly repetitive over-reliance on comparative language techniques like similes and metaphors as well as stilted-sounding dialogue and prose which was said to be poorly written.
"In response to the accusations, Ballard confessed that she herself had not used AI when writing Shy Girl but that a third-party whom she’d hired to edit the self-published edition of her novel had."
As concern from readers began to grow, Shy Girl was ran through a AI-detection software by the name of Pangram which reported the novel was 78% AI-generated. As a result, the US release of Shy Girl was cancelled and the novel was withdrawn from publication in the UK. In response to the accusations, Ballard confessed that she herself had not used AI when writing Shy Girl but that a third-party whom she’d hired to edit the self-published edition of her novel had. By this point, however, the damage was already done in terms of tarnishing the reputations of both Hachette and Ballard herself, with readers questioning how a book whose AI-modification had been so seemingly evident could have made it to the publishing stage without questions being raised and calling for more clear-cut regulations to ensure authors report any use of AI within the writing process or to what extent it should even be permitted.
Though Shy Girl is the first book to be dropped by a major publisher for its reported use of AI, it is not the first time authors have referenced using AI, nor is it likely to be the last. American author James Frey has openly admitted to using extensive amounts of generative AI to assist his writing process, yet he still continues to be published and maintains a moderate fan base. Comparing this to the fact that the reported use of AI in Shy Girl has led to an aggressive barrage of online hate directed towards Ballard, to the extent she has taken herself entirely off social media and claimed her name has been “ruined”, the comparison in responses to Ballard and Frey becomes a deeply complex symptom of the poor regulation of AI use within the publishing industry, something which must be addressed sooner rather than later.
"Storytelling is an ultimately human act and, as readers, we put our trust in authors, along with their surrounding networks of editors and publishers, to deliver deeply human narratives..."
In my opinion, the use of AI within all creative spaces, literature primarily included, is fundamentally wrong, especially since novels are meant to be original works of innovation which spark joy in readers and champion creativity. Storytelling is an ultimately human act and, as readers, we put our trust in authors, along with their surrounding networks of editors and publishers, to deliver deeply human narratives within a publishing environment that has an unfaltering commitment to authenticity and integrity. But unlike visual creative mediums where AI alterations are easier to spot, attempting to regulate the use of AI within literature is borderline impossible. No one can be a 100% accurate AI-detector - even software that is used to detect AI like Pangram is not right all the time - but if we cannot trust the very establishments within publishing that claim to have a commitment to the truth and championing original works yet make still allow such oversights, who can we trust?
The first and ultimately best step forward would be employing a firm stance across the entirety of publishing about how tolerated the use of generative AI is within literature, where it is either permitted under strict guidelines or banned entirely. As much as a wide proportion of readers - myself included - would like to see staunch advocacy from the publishing industry against generative AI use to promote and protect original works in creative fields, steps must be taken on an institutional level either way to ensure there is no repeat of the Shy Girl controversy and that authors, publishers and novels alike are not caught in the crossfire of such harmful miscommunications surrounding AI use again.