Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 39 of the 39 references it cites were found in a registry or an index; 30 of the 58 sources its claims cite were checked for support.

Playwriting with Large Language Models: Perceived Features, Interaction Strategies and Outcomes

Open access
Grigis, Paolo, De Angeli, Antonella · 2024 · Proceedings of the 2024 International Conference on Advanced Visual Interfaces

Abstract

Large Language Models (LLMs) are sparking debates about creativity, intellectual property, and artistic integrity. This paper focuses on creativity, defined as consensual agreement among domain experts. It presents an inductive analysis of seven semi-structured interviews with professional playwrights who engaged in a longitudinal project with the aim of writing a theatre script using commercial systems. Overall, participants regarded LLMs as unsuitable for playwrighting. However, they enjoyed the experience and identified utility for editorial tasks and brainstorming. A significant obstacle was associated with the politics embedded in LLMs. Not only did these systems avoid a language that could offend sensibilities, but they also refused to engage in taboos and conflicts, which are the core of dramaturgy. Other system features (speed, exploitation, and unpredictability) were sometimes considered conducive and sometimes detrimental to creativity. Participants experienced difficulties and tried to build common ground by trial and error. Often, this strategy evolved into role play: the playwright instructed the LLM to enact characters. The interaction provided hints of inspiration and fostered suspension of disbelief and ontological reflection. However, it often led to technology rejection. Comparing and contrasting our insights with related work, we conclude by opening new directions for research at the boundaries of HCI and AI.

Overview

CoChat Summary

TL;DR

An inductive interview study with seven professional Italian playwrights in a seven-month project using commercial LLMs found they regarded LLMs as unsuitable for playwriting, though they enjoyed the experience and saw utility for editorial tasks and brainstorming. Roleplay and trial-and-error interactions produced hints of inspiration and ontological reflection, but embedded politics, taboos, and other system features often led to technology rejection.

Key findings

  • Participants overall regarded LLMs as unsuitable for playwriting, but identified utility for editorial tasks and brainstorming; a major obstacle was the politics embedded in LLMs, which avoided offensive language and refused taboos and conflicts described as core to dramaturgy.
  • Perceived features were speed, exploitation, politics, and unpredictability: speed enabled large output but was linked to “low-quality scripts,” superficiality, and loss of critical thinking; exploitation included using the system as an “artificial assistant,” “slave,” or “butler,” while some feared the machine exploiting their work, with P7 refusing to fine-tune the model with their previous work.
  • Interaction strategies were trial-and-error prompt writing and roleplay; roleplay supported suspension of disbelief, hints of inspiration, and ontological reflection, but often led to technology rejection.
  • All playwrights perceived the “creative effort” as uniquely their own; the machine could not structure narratives, and some stopped using LLMs due to lack of time/context, disconnection, or interactions devolving into clichés and stereotypes.

Methods

  • Seven semi-structured interviews were conducted in June 2023 with seven professional Italian playwrights aged 25–53 (mean age 36), all with backgrounds in playwriting and acting, after a seven-month education project (Nexus 23/24, December 2022–June 2023) using GPT-3/3.5 tools SudoWrite, Playground, and ChatGPT.
  • Interviews were online, lasted one hour, were audio recorded and manually transcribed; analysis used the general inductive method, producing three high-order themes, with an external researcher independently coding 30% of the data and Cohen’s Kappa surpassing 0.80 for each low-level theme.
  • In June 2023, all but one of the artists enacted selected scripts in a final play with traditional acting and improvisation elements; the main topic was artificial intelligence itself.

Limitations

  • Stated: the study involved a small group of Italian playwrights engaged in collective work over seven months, so mutual influences are expected; commercial LLMs all derived from the same foundation model, due to a lack of specialised systems in Italian; participants often acknowledged a lack of technical knowledge as a personal constraint.
  • Evident from the text: all participants were professional playwrights/actors with no prior experience using LLMs in their professional practices, so findings may not generalise to technical experts or other creative domains; interviews were conducted after the performance and quotes were translated from Italian, with gender pronoun handling noted.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1145/3656650.3656688

Cites

39 of 39 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Thematic analysis results.
Figure 1 Thematic analysis results.CC-BY · From the PDF

Verification

Claims the paper makes

15 Valid · 2 Not valid · 14 Not checked · 111 Own results and interpretations

53 more claims · 1 with their cited sources checked

Create a free account to see all 142 claims and to verify the 4 references still open.

See all claims — free account
  • It may state two separate claims as one. 44
  • Probably not a substantial claim. 61
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 37 references

Research works printed but not linked: 11

  1. Teresa M. Amabile. 1988. A model of Creativity and Innovation in Organizations. Research in organizational behavior 10, 1 (1988), 123–167.
  2. Chris Bilton and Stephen Cummings. 2014. Handbook of Management and Creativity (1 ed.). Edward Elgar Publishing, Cheltenham, UK.
  3. 10.5555/3495724.3495883 · doi:10.5555/3495724.3495883
  4. Alex Calderwood, Vivian Qiu, Katy Ilonka Gero, and Lydia B. Chilton. 2020. How Novelists Use Generative Language Models: An Exploratory User Study. In Proceedings of IUI ’20 workshops, Mar 17-20, 2020 (Cagliari, Italy). Association for Computing Machinery, New York, USA, 5 pages.
  5. Antonella De Angeli, Graham I Johnson, and Lynne Coventry. 2001. The Unfriendly User: Exploring Social Reactions to Chatterbots. In Proceedings of the international conference on affective human factors design, June 27-29, 2001 (Singapore, Japan). Asean Academic Press Ltd, New Orleans, USA, 467–474.
  6. Antonella De Angeli, Paula Lynch, and Graham Johnson. 2001. Personifying the E-Market: A Framework for Social Agents. In Proceedings of Human-Computer Interaction : IFIP TC13 International Conference on Human-Computer Interaction, July 9-13,2001(INTERACT ’01). IOS Press, Tokyo, Japan, 198–205.
  7. Isabelle Gatt. 2020. Theatre: a space for human connection. Taboo: The Journal of Culture and Education 19, 2 (2020), 5.
  8. Bella Poynton. 2020. Playwriting with Robots: Creating Cyborgian Theatre with Online Chatbots. Theatre/Practice: The Online Journal of the Practice/Production Symposium 9 (2020), 1–31.
  9. Sandra W. Russ and Claire E. Wallace. 2013. Pretend Play and Creative Processes. American journal of play 6, 1 (2013), 136–148. Publisher: ERIC.
  10. Katie Salen Tekinbas and Eric Zimmerman. 2003. Rules of Play: Game Design Fundamentals (1 ed.). MIT press, London, UK.
  11. Langdon Winner. 1980. Do Artifacts Have Politics?Daedalus 109, 1 (1980), 121–136.

Cited by

See all 14

Metadata

Authors
Grigis, P., & De Angeli, A.
Affiliations
  • Free University of Bozen-Bolzano (IT)
  • Free University of Bozen/Bolzano, Italy
Venue
Proceedings of the 2024 International Conference on Advanced Visual Interfaces
Publisher
ACM
Published
2024-06-03
Pages
1-9
OpenAlex
W4399203197
Cited by
13
References
37
Citations by year
2026: 4 · 2025: 7 · 2024: 2
FWCI
45.65
Type
conference paper
Language
en
Open access
Locations
  • International Conference on Advanced Visual Interfaces (AVI) (publishedVersion, cc-by)
  • Open Research (University of Surrey) (submittedVersion, cc-by)
Funders
European Commission
Topics
  • Social Sciences › Arts and Humanities › Visual Arts and Performing Arts › Art, Technology, and Culture
  • Social Sciences › Arts and Humanities › Visual Arts and Performing Arts › Artistic and Creative Research
  • Social Sciences › Arts and Humanities › History and Philosophy of Science › Cybernetics and Technology in Society
Keywords
large language models, playwriting, creativity, AI-assisted writing, brainstorming, dramaturgy, role play, suspension of disbelief, resistance to technology, human-AI interaction

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Playwriting with Large Language Models: Perceived Features, Interaction Strategies and Outcomes. https://app.cochat.ai/papers/10.1145/3656650.3656688
The paper
Grigis, P., & De Angeli, A. (2024). Playwriting with Large Language Models: Perceived Features, Interaction Strategies and Outcomes. Proceedings of the 2024 International Conference on Advanced Visual Interfaces. https://doi.org/10.1145/3656650.3656688

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.