From Homer's Odyssey to Agatha Christie's crime fiction, humans have long stocked the bookshelf. New research suggests artificial intelligence is now providing some competition — and, by readers' ratings, winning.
In a study published in the journal Judgment and Decision Making, 1,682 adults each read one of six short stories: three written by humans and three by ChatGPT, with each AI story sharing a theme with a human-authored counterpart. Participants were told whether their story was written by a human or an AI — but that information was not always accurate. They then rated how absorbing and engaging the story was, and its quality.
Readers who actually received an AI-generated story rated it as more absorbing and of higher quality than those who read a human-written story. Yet stories labeled as human-written earned higher ratings regardless of their true origin — a preference the researchers link to a desire for authenticity.
"AI systems can already generate short stories that are seen as at least as good as — if not better than — human-written stories," said Dr. Deena Skolnick Weisberg, a senior author of the research from Villanova University. "We should update our views of AI's abilities accordingly." Weisberg, herself a creative writer, stressed the finding does not make human writers obsolete: "The fact that AI can generate human-like stories doesn't change any of that" — the value of writing as self-expression, challenge, and making sense of experience.
In two follow-up experiments with 905 adults asked to pick the human story from a pair, only about 40% guessed correctly in one experiment — worse than chance — and 52% in the other. "There is no general evidence that participants were able to distinguish human-written from AI-generated stories," the authors write. Greater experience with AI systems improved accuracy, while self-reported fiction expertise did not. Not everyone welcomes the result: Luke Kennard, a creative writing professor at the University of Birmingham who was not involved in the work, called AI "an existential threat," arguing its output rests on training data drawn from human authors and on resource-intensive data centers — "is it worth it? And the answer is absolutely, categorically, it is not."
