How Do Human Creators Embrace Human-AI Co-Creation? A Perspective on Human Agency of Screenwriters

TL;DR

This paper investigates how professional screenwriters exercise agency in human–AI co-creation, using Bandura's four properties of human agency (intentionality, forethought, self-reactiveness, self-reflectiveness) as an analytical lens. Through a two-week qualitative study with 19 participants, employing co-creation sessions, retrospective think-alouds, and semi-structured interviews, the authors reveal how writers form plans, anticipate AI outcomes, evaluate and regulate AI outputs, and iteratively develop new workflows, strategies, and cognitive approaches. They identify three core contributions: a developmental account of how screenwriters cultivate co-creation practices and self-improvement through sustained AI interaction; a set of future AI system requirements framed as agency-enhancing roles (plan-making partner, proactive thinker, gatekeeper, reflective mentor); and practical guidelines for screenwriters, tool designers, and researchers to sustain human agency in long-term human–AI collaboration. The findings highlight opportunities and risks in co-creation, including skill development, confidence, social and emotional dimensions, and the potential for over-reliance or skill erosion, underscoring the need for longitudinal, domain-diverse studies and thoughtfully designed tools that support deliberate, value-driven human creativity.

Abstract

Generative AI has greatly transformed creative work in various domains, such as screenwriting. To understand this transformation, prior research often focused on capturing a snapshot of human-AI co-creation practice at a specific moment, with less attention to how humans mobilize, regulate, and reflect to form the practice gradually. Motivated by Bandura's theory of human agency, we conducted a two-week study with 19 professional screenwriters to investigate how they embraced AI in their creation process. Our findings revealed that screenwriters not only mindfully planned, foresaw, and responded to AI usage, but, more importantly, through reflections on practice, they developed themselves and human-AI co-creation paradigms, such as cognition, strategies, and workflows. They also expressed various expectations for how future AI should better support their agency. Based on our findings, we conclude this paper with extensive discussion and actionable suggestions to screenwriters, tool developers, and researchers for sustainable human-AI co-creation.

Figures (6)

• Figure 1: Research Overview. This figure first presents how we link the four properties of human agency to the study's three research questions. It also briefly summarized the research method to address our research questions and overall findings from our study.
• Figure 2: CloseChat Platform Interface. The interface illustrates the experimental environment, including the available system features and an example dialogue. The left side of the figure shows the full platform layout, and the right side presents a partial example of the model's output from one interaction round.
• Figure 3: Findings for RQ1. This figure illustrates a sequential process through which screenwriters enact agency when co-creating with AI. Screenwriters first form plans, then anticipate outcomes, and finally react to AI outputs through immediate evaluation and response.
• Figure 4: Findings for RQ2. This figure summarizes detailed findings across six aspects of screenwriters' development and adaptation grounded in self-reflection, including strategy development, confidence enhancement, workflow reorganization, concerns about skill sets, cognitive change, and reconsideration of AI roles.
• Figure 5: Findings for RQ3. This figure maps participants' envisioned AI roles onto four human agency-oriented dimensions: intentionality, forethought, self-reactiveness, and self-reflectiveness. It illustrates screenwriters' expectations for how future AI systems may support different aspects of human agency in co-creation, including acting as plan-making partners to support intentionality, as proactive and farsighted thinkers to enhance forethought, as responsible gatekeepers for self-reactiveness, and as reflective mentors to facilitate self-reflection.