Automating Artistry: The Social Impact of Generative AI on Creative Labor in Theatre

Abstract

The paper examines the use of generative AI in the production of stage plays and its impact on classical forms of drama. As AI takes on a greater role in creative processes, one asks if screenplays written by machines can capture the structural and emotional dimensions of theatrical narrative. To solve this problem, we propose a novel model named Dramatic Structure-Aware Multi-Agent Generative Framework (DSA-MAG). The model consists of a transformer-based encoder, multi-agent narrative generation, dramatic structure analyzer, emotional coherence module, and character consistency optimizer. In this study, we use the dataset of Movie Scripts Corpus collected from Kaggle. The dataset contains 1200 script segments from various genres. The segments were vectorized using a HashingVectorizer with 1500 features. The experimental results showed that the DSA-MAG model achieved the best accuracy of 25.40%. The metrics of precision 0.0788, recall 0.2540 and F1-score 0.1179 further validated the superiority of the model in multi-class genre categorization. Therefore, these results are consistent with the claim that the more unified and more fundamentally perceptive approaches to script generation with structured multi-agent AI frameworks improve AI-assisted creative writing as well as research on theatrical drama.