A short story widely accused on social media of being generated by artificial intelligence has won the overall Commonwealth short story prize, according to The Guardian.
The winning piece, titled The Serpent in the Grove by Jamir Nazir, went viral after being announced as a regional winner in mid-May.
Critics on X and Bluesky pointed to stylistic features they claimed indicated machine origin.
Following the backlash, literary magazine Granta withdrew from its long-standing partnership to publish the winning entries.
The Commonwealth Foundation conducted an internal review of regional winners, evaluating drafts, time-stamped documents, and author notes to verify authenticity.
“We are satisfied with the testimonies of our writers and their confirmation that AI was not used in their writing,” said foundation director-general Razmi Farook.
With this overall victory, Nazir receives an additional £2,500 prize, on top of the £2,500 he secured for winning the Caribbean regional category last month.
Judging chair Louise Doughty praised the selection, describing the text as “an original, poetic and deeply moving story”.
Online critics highlighted specific recurring linguistic patterns, such as multiple “not x, but y” structures and lists of three, as evidence of automated generation.
Skeptics also called attention to specific lines in the narrative, including: “Sun on galvanise is a cruel instrument”; “She had the kind of walking that made benches become men”; and “Marsha lived two bends down … [she was] big in the way of women who never apologise to furniture”.
In a promotional film released by the organizers, Nazir cited authors VS Naipaul and Derek Walcott as major creative influences.
He explained that he produced six or seven drafts, using speech-to-text software on his mobile phone.
Because the screen only displayed a few lines at once, he meticulously polished each line before proceeding.
“This story began in my childhood in rural Trinidad,” he said. “Each day, I walked to school past rum shops where cane workers and labourers gathered.
I remember the voices, the laughter, the arguments and conversations that shaped village life.”
Initial public reception to the announcement remained largely critical.
Some social media users disclosed that they had processed the text through various AI-detection programs during the initial regional selection phase.
“Pangram flags at 100% but also, come on, if you know you know”, said Wharton professor Ethan Mollick.
However, the reliability of AI-detection software has been called into question.
Farook defended the organization’s decision to rely on human assessment rather than automated screening tools during the verification process.
“Rather than surrender our judgment to AI-detection software, we asked our winners to show their working drafts, outlines, the evidence of an artistic journey.
That software, it must be said, is not infallible: it returns inconsistent verdicts and, in doing so, corrodes the very trust on which a prize depends.”
“When the machine’s default voice is the metropolitan one, the writer who does not fit the expected mould is the first to fall under suspicion,” she added.
“The more startling her gift, the more her unfamiliar brilliance unsettles, the more readily she is accused of being a machine.
A young writer in Kingston or Kolkata, in Kuala Lumpur or Kigali, must now prove not only her talent but her very humanity.”
