generated tales preferred to human
A study from Villanova University found that readers often prefer AI-generated stories over human-written ones when they believe the AI content was written by a person. Participants struggled to distinguish between the two, suggesting a bias toward human authorship.
Why it matters
This research challenges public assumptions about the creative limitations of AI and highlights how human perception influences the evaluation of creative work.
Share Save Add as preferred on Google Aimee Dexter Cambridgeshire EPA/Shutterstock Researchers used three fictional short stories written by published authors, alongside three generated by ChatGPT Readers gave the highest ratings to stories created by artificial intelligence (AI), believing they had been written by humans, according to new research.
The study, published by Cambridge University Press, tested to see whether people aged 18 to 81 could tell the difference between human-written and AI-generated short stories.
Led by researchers at Villanova University (VU), USA, it found that participants could not tell which was which.
Senior author Dr Deena Weisberg, from the department of psychological and brain sciences at VU, said: "I'd say that public assumptions about AI's capabilities are increasingly out of date."
The study also examined how participants rated the appeal and quality of each story and whether they could identify AI-generated content.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in