AI at the greenlight: negotiating creative agency in early-stage film development

The growing use of artificial intelligence (AI) in screenwriting and film development has intensified debates around creativity, authorship, and the role of algorithmic systems in creative processes. AI-driven analytics systems now evaluate screenplays to forecast box-office performance, model audience demographics, and inform casting choices. While scholarship on AI tools in the creative industries has expanded rapidly in recent years, studies focusing on the screen industries – particularly on AI-driven analytics systems used in screenplay evaluation – especially from a practice-led perspective, are still scarce. This study contributes to this area by presenting a practice-led case study that observed the use of a Swiss-based AI-driven analytics system in the development of a German – New Zealand feature film, Come Together. Drawing on academic literature, industry reports, and the author’s involvement in the script development of Come Together, this article discusses the ontological tensions between human creatives and algorithmic evaluation. By examining these tensions, the article develops three interconnected conceptual insights – creative flattening, risk of formulaic feedback, and blind spots – as a framework for understanding the negotiation of creative agency when algorithmic evaluation enters the early stages of film development.