Who Tells the Better Story? Comparing Human and AI-Generated Narratives on Perinatal Mental Health Topics

Storytelling is a powerful tool for supporting perinatal mental health, allowing mothers process experiences, reduce isolation, and foster connection. Although generative AI is increasingly used in therapeutic and peer-support interventions, it remains unclear how AI-generated stories are perceived compared to narratives grounded in lived experience. We conducted a mixed-methods study to investigate the perceptions of AI-generated and human-authored stories. Using MTurk, we collected 400 responses from female participants in their perinatal stages. Participants rated stories on believability, engagement, relevance, and emotional catharsis. Results showed that AI-generated stories were often described as polished and coherent but emotionally distant, while human-authored stories, adapted from online communities, were viewed as raw, culturally specific, and emotionally resonant. Emotional catharsis was tied less to positivity than to the progression of emotions. These findings underscore the continued centrality of lived-experience narratives in perinatal support and suggest that any use of AI storytelling in this domain should be cautious, transparently bounded, and clearly secondary. We discuss implications for designing human-centered storytelling technologies that preserve authenticity while carefully delimiting the role of AI in perinatal mental health contexts.