A new study led by UC Riverside professor Demetrius Lewis reveals that applying modern genre labels to historical works often creates a retrospective bias that distorts how art is understood and valued. Published in the journal Academy of Management Discoveries, the research argues that genre classifications change over time, and using contemporary definitions to analyze past works can lead to systematic misunderstandings of creative intent and audience reception. The film "Sinners," for example, is often categorized as a horror movie due to its gory vampire attacks. However, the creators used horror elements as a metaphor for violent racism in the early 20th-century American South, making it equally a period drama. When such works are mislabeled by current standards, they may be penalized for spanning multiple genres, a trait that often signals creativity but can confuse audiences expecting traditional category boundaries. This misclassification can result in lower ratings and hinder discovery, as streaming platforms and marketers rely on genre tags to recommend content. Lewis and his co-authors, Giacomo Negro of Emory University and Isin Guler of the University of North Carolina at Chapel Hill, propose that genres evolve by absorbing innovation. Historical examples include the blending of acoustic folk and electric rock in the 1960s to create folk-rock, and the 1976 film "Carrie," which expanded the horror genre to include themes of puberty and social rejection. As these styles become familiar, the boundaries of their respective genres expand. However, analyzing such works with rigid, modern labels obscures these evolutionary relationships. To address this, the team developed an artificial intelligence tool that uses large language models to reconstruct the genre classifications of cultural products as they were originally understood. The system prompts the AI to assign labels based on two scenarios: one using current knowledge and another restricted to the cultural context of the work's release year. By comparing these two datasets, the tool identifies how genre definitions have shifted and translates past categories into their modern equivalents. This process effectively harmonizes historical and contemporary classification systems, providing a nuanced view of how a film or album would have been perceived at its debut. This AI-driven approach offers significant benefits for both creators and audiences. For creators, it reduces the risk that boundary-pushing projects will be undervalued or misidentified because they do not fit neatly into established boxes today. For audiences, it improves the accuracy of content discovery and interpretation, ensuring that viewers understand the full scope and original intent of a work rather than being misled by oversimplified or anachronistic labels. The study emphasizes that while the penalty for genre ambiguity is not fixed and often diminishes as audiences adapt, correcting retrospective bias allows for a fairer assessment of innovation. By capturing how creative products were actually experienced when first released, the tool helps balance the evolution of genres with the need for accurate historical analysis. Ultimately, the goal is to provide a clearer picture of artistic identity, ensuring that creative works are judged by the standards of their time rather than the potentially distorted lens of the present.