Thélyson Orélien relied on artificial intelligence to pen his bestselling, prize-winning debut novel.
The controversy erupted just hours after the 38-year-old author was officially awarded the highly coveted Prix du Roman Fnac for his critically acclaimed migration story, C’était ça ou mourir (It Was Either That or Die). A newly surfaced digital watchdog account on X, running under the handle Balance ton Claude (Denounce your Claude), shocked the publishing world by claiming that key passages of the book flagged as a staggering 100% AI-generated when analyzed by the Pangram AI detection platform.
Orélien’s novel has been hailed as the absolute phenomenon of France’s autumn literary season, racking up over 35,000 copy sales within its first five weeks. It has also earned coveted placements on the initial longlists for some of the world’s most prestigious French-language literary honors, including the Académie Goncourt prize.
The author has vehemently rejected the accusations, branding the automated findings a direct, malicious attack on his creative integrity. Speaking to the French daily Libération, Orélien expressed deep anger over the claims. ‘I have always loved creation and literature. I can’t see why I would feel compelled to use a machine,’ he stated, adding that his lyrical prose, repetitive imagery, and distinct rhythms are drawn purely from his Haitian and Caribbean roots.
As part of his defense timeline, Orélien noted that the conceptual genesis of the book began in 2017, with an initial manuscript already drafted by 2019—years before public generative AI platforms became commercial staples.
Despite the growing online uproar, the author’s publishers—Grasset in France and Éditions du Boréal in Canada—have issued fierce joint statements confirming they stand solidly behind Orélien. Grasset dismissed the automated text checks as ‘atrociously injurious,’ emphasizing that neither house has any plans to suspend distribution or withdraw the book from store shelves.
The unfolding scandal has ignited an intense global debate over the limitations of AI-detection algorithms, racial bias in automated policing, and how literary committees will safeguard original human authorship moving forward.
2026-09-24
