The Commonwealth Prize controversy shows that the literary world is not ready for AI

Since 2012, British magazine Grant has published the regional winners of the annual Commonwealth Short Story Prize. This year, however, there was something about one of the nominations for the top prize: It appears to have been written by AI.

It’s Jamir Nazir “Snake in the Bush” he has many signs of prose created by LLM – mixed metaphors, anaphora, series of three. (I know this, too, is a series of three, and I promise that I wrote this myself, without help, while writing all the things.) I admit that at first I did not believe that Nazir’s story was created by AI. I know people are using LLMs to help them write — or write for them, period — but I’ve been wary of the kind of AI paranoia that’s grown among my peers. Em dashes are said to be AI telling, as are the words “search” and the aforementioned lists. Short, punchy sentences, too, especially when used to refer to long sentences in a row.

But I, man, have already used all of these in my writings. LLMs, after all, are taught in humanities. They reflect what they were fed. And yet there is an eerie quality to AI-generated prose. There is something about this, even if you can’t tell right away what it is. If there are AIs telling us, and I’m using this word right now, then how do you know I wrote this?

Nabeel S. Qureshi, a former visiting AI expert at the Mercatus Center at George Mason University, was among the first to describe the use of AI in Nazir’s article. For Qureshi, the first two sentences were proof enough.

They say that the forest still cries during the day. Not a good bee trade or a clean cutlass wipe on the vibe, but a belly rumble – as if the earth swallowed the cry and kept it there..

“In general, AI writing has a tone that I learned that is hard to describe,” Qureshi told me via email. “There is a trend from ‘AI helped me change’ to ‘AI wrote this’ – this case counts to me as the end of it, although I don’t know for sure.

The problem is that although the use of AI is widely suspected, no one really knows. In a statement, Commonwealth Foundation CEO Razmi Farook said the organization is aware of the AI ​​claims in successful stories, including Nazir’s. Farook said that all writers who submitted work for the award were asked if they were submitting original, unpublished articles, and that all the selected writers said that no AI was used to help them write their stories.

“Until a comprehensive tool or method for identifying the use of AI emerges that can also deal with the challenges associated with unpublished fiction, the Foundation and the Commonwealth Short Story Prize must operate on a principle of trust,” Farook said.

Grantfor his part, he published Nazir’s story through Claude “and asked if it was made by AI,” publisher Sigrid Rausing. he said in his voice. “The response was long, thinking ‘it’s not automated.'” But Claude isn’t an AI detection tool, it’s a chatbot powered by a big language. Although AI tools are often better than human readers at recognizing prose produced by LLMs – or those who judge literary awards – GrantWord shows that he went there to ask if the article was actually created by AI, which suggests that the magazine may not understand how AI works.

“It may be that the jury is now going to award an AI fraud case — we don’t know, and we probably never will,” Rausing said.

The media is booming tricked to start AI-generated storiessome of them are “written” by “writers” who do not really exist. There were also doubts that Nazir himself was a fake – even the writer Kevin Jared Hosein, winner of the Commonwealth Short Story Prize, confirmed that Nazir was a real person, and he shared the messages he shared recently and Nazir on AI suspicions use his story. Nazir as well published poetry in 2018.) Nazir did not respond Seasiderequest for comments. In March, Hachette he pulled out a flyer about Mia Ballard’s horror novel Shy Girl after being accused of using AI, although Ballard denied using it and criticized the hiring editor.

There is also the question of whether there is any legitimate way for writers and journalists to use AI. Prose produced by an LLM is obviously verboten, but what about using AI to generate ideas, or research? What about AI transcription services? Does relying on these tools mean that the job is no longer yours? This week, Polish author Olga Tokarczuk he agreed he uses AI to support his creation – the other end of the AI-use spectrum Qureshi mentioned, but a shock to readers who admired the author who won the Nobel Prize in Literature.

“I often throw an idea into the machine and ask: ‘Honey, how can we explain this better?'” says Tokarczuk, who received the top literary honor in 2018:

“Although I am aware of his magic and many errors in terms of economics or reality, I must admit that in fiction, this technique is of great value. For all these reasons, I am very sorry for Balzac, Cioran, and Nabokov, because although I am interested, I do not believe that modern chats can speak well.

Tokarczuk’s comments, delivered in Polish at a recent ceremony in Poznań, had the misfortune of going viral at the same time as the Commonwealth Prize debate. (We translated his words into English with a human.) But he’s more against AI than the headlines surrounding the event. Tokarczuk explained how he uses AI in a three words shared with Lit Hub in which he explained that he didn’t use AI to write the upcoming book but instead used it to “quickly write and check facts,” although he personally verified the content.

“Sometimes I am inspired by dreams,” he continued, “but before this sentence is closed and torn apart by experts, I hasten to say that it is my dream.”

The tone of Tokarczuk’s initial comments – and the need he felt to explain himself – speaks to the overarching, irrational nature of the public’s use of AI. Prose produced by an LLM may be innovative, but is it what everyone wants? Thousands of people threatened to boycott Barnes & Noble after CEO James Daunt said he has no problem selling books written by AI, as long as the books have disclaimers that they weren’t written by a human. Daunt later responded to his comments, but not completely. “Book banning is a clear and present danger, so we take great care to ban any books,” he said Los Angeles Timesand also ensure that they “don’t sell AI-generated books that look like real authors.”

None of this, however, explains the uniqueness of AI-generated work, or what distinguishes bad LLM-generated writing from human-generated bad writing. When I ran Nazir’s story through Pangram, an AI- and fraud-detecting program, it came back as 100 percent AI-generated. According to Pangram, the most obvious issues were Nazir’s use of three triads; the word “stubborn,” which occurs six times more in AI-generated text than human-generated text; and the word “if he had,” whose appearance is five times more likely. But here we have another series of three, written by me, the man.

Unsatisfied, I released the unpublished text in my upcoming book, which I am editing, through Pangram. Only one paragraph was included two three. (It’s not the best part of the book, which is why I’m editing it.) Pangram said that part was 100 percent human-written, which is true, but I’m still not convinced. I took another part out – a better one, I think – and it said the same thing. When I ran the first chapter of Edge book by editor Kevin Nguyen, In Documentsthrough Pangram, the results were the same. Pangram himself ran every Commonwealth Prize winner through his programs, and he found out two of the 2026 award winners, as well as the 2025 winner, appear to have been created by AI. Human-made work has an incredible quality, just as it changes. Maybe AI-generated prose is like junk: You know it when you see it, even if you don’t know why.