Haitian-Canadian writer Thélyson Orélien has denied allegations that he used artificial intelligence to write most of his debut novel, C’était ça ou mourir (It Was Either That or Die).
Orélien, 38, recently won the Prix du Roman Fnac, a major French literary award, for the novel. However, the recognition has been overshadowed by claims that artificial intelligence was used to generate the work.
The allegation was made by an X account, Balance ton Claude, which said it subjected passages from different sections of the novel to an AI-detection platform known as Pangram.
The account claimed that the passages returned what it described as a “100 per cent AI” result.
Orélien rejected the allegation, saying he had no reason to rely on a machine to produce his literary work.
“I have always loved creation and literature. I can’t see why I would feel compelled to use a machine,” he told French newspaper Libération, according to the BBC.
The author said he was working with his French and Canadian publishers to compile evidence showing that he personally wrote the novel.
“My traditions are Haitian and Caribbean. The images, the repetitions and the rhythms are the living word, they are my own,” Orélien said.
His French publisher, Grasset, also defended the writer, describing the allegations as damaging to an author with an “exceptional life-story” and maintaining that his work was original.
The controversy comes as Orélien’s novel continues to receive attention in France. The book has been shortlisted for the Goncourt and Renaudot literary prizes and has attracted interest beyond the country.
The novel tells the story of a teacher who escapes gang violence in Haiti and eventually travels to Canada, where he finds a life free from the constant fear he experienced in his home country.
Orélien was born in Gonaïves, Haiti, in 1988 and moved to Montreal in 2010. He later worked as a journalist in Canada.
According to the author, the first draft of the novel was completed in 2019, before generative AI writing tools became widely accessible.
The dispute has also renewed debate over the reliability of AI-detection systems and whether their results can conclusively establish that a piece of writing was produced by artificial intelligence.
