Researchers at the University of North Carolina at Chapel Hill found that while artificial intelligence can spin increasingly convincing stories, its characters may still lack mystery. The study examined how characters in stories generated by AI compare with those written by people, analyzing eight different aspects of character portrayal. The team developed CASPER, an automated framework that evaluated thousands of stories. "We found that AI models tend to 'play it safe' with their characters, in the sense that they wrap up storylines neatly," said Anneliese Brei, a graduate student in computer science at UNC-Chapel Hill and lead author. "Human writers, on the other hand, are sometimes more willing to leave questions unanswered and let characters remain mysterious." The analysis revealed that AI-generated characters often lean more heavily on recognizable archetypes and tend to arrive at tidy resolutions. "One of our most surprising findings was that bigger and more powerful AI models don't necessarily create more varied characters than smaller ones," said Nicholas Sanaie, an undergraduate student in computer science at Carolina and co-author. CASPER gives researchers, developers and creative professionals a way to benchmark whether newer AI systems are actually improving portraying complex characters.