CoChat
Sign Up
Cited by 13Artistic and Creative ResearchVirtual Reality Applications and Impacts

wr-AI-ter: Enhancing Ownership Perception in AI-Driven Script Writing

Open access
Weber, Christoph Johannes, Burgkart, Sebastian, Rothe, Sylvia·2024·ACM International Conference on Interactive Media Experiences
DOI 10.1145/3639701.3656325
Open this paper in CoChat — free accountOpen PDF

Abstract

The integration of artificial intelligence (AI) into creative domains is increasing, presenting both challenges and opportunities. In screenwriting, personal artistic expression is a fundamental aspect of the creator’s identity and work. The current use of AI in such creative processes can sometimes overshadow the creator’s vision and lead to a reduced sense of ownership over the final product. We introduce wr-AI-ter, an interactive application consisting of four basic stages: Ideation, Structure, Refinement, and Export. While some related work focuses on experts The application is intended to aid users with varying levels of screenwriting proficiency in generating screenplays using artificial intelligence, while preserving their sense of authorship. We conducted a user study with 23 participants, who had different expertise (screenwriting, documentary filmmaking, and VFX artistry). The results indicate that AI has the potential to accelerate the screenwriting process and improve the quality of scripts without compromising the sense of ownership.

AI snapshot

Written from the paper’s full text.

TL;DR

The paper introduces wr-AI-ter, a four-stage AI-assisted screenwriting app built with GPT-4, and tests whether it can support script creation while preserving users’ ownership. In a 23-person mixed-expertise study, most users found it helpful and enjoyable, and 57% reported ownership, though uniqueness, surprise, and expression scores were low.

Key findings

  • Ownership was reported by 57% of participants (43% agreed, 13% strongly agreed). VFX artists showed the highest agreement at 80%, followed by screenwriters at 78%; documentary filmmakers largely reported a lack of ownership at 78% (67% disagreed, 11% strongly disagreed).
  • Helpfulness and enjoyment were high: 71% endorsed helpfulness (28% agreed, 43% strongly agreed) and 71% enjoyed using the tool (42% agreed, 28% strongly agreed). Ease of use was affirmed by 48%, and collaboration by 43%.
  • Weak areas included expression, uniqueness, and surprise: only 21% agreed they could express creative goals, 14% found outputs unique, and 35% were surprised. Pride was modest, reported as 39% in the results section (26% agreed, 13% strongly agreed), though the discussion later cites 46%.
  • Free-form responses praised brainstorming and plotting support but criticized generic output, limited AI responsiveness, insufficient character-development tools, and weak integration with final scriptwriting. Compared with Dramatron, wr-AI-ter had higher ownership but lower pride, ease, expression, collaboration, uniqueness, enjoyment, and surprise (e.g., uniqueness 14% vs. 69%; surprise 35% vs. 92%).

Methods

  • The authors built an interactive application with four stages—Ideation, Structure, Refinement, and Export—based on Freytag’s five-act structure. It includes AI features for story-beat suggestions, question-and-answer analysis, and emotion enhancement, implemented in Swift/SwiftUI with GPT-4 via the OpenAI API, and supports export to Fountain and treatment formats.
  • A user study recruited 23 participants aged 18–39 from screenwriting (P1–P9), documentary filmmaking (P10–P18), and VFX artistry (P19–P23). The sample included 13% experts, 17% advanced or moderately experienced writers, and 70% beginners; 83% had prior experience with AI writing tools. The study was bilingual in German and English, and participants spent about 30 minutes composing at least one screenplay.
  • Participants rated nine statements on 5-point Likert-type scales adapted from Dramatron and Wordcraft, covering ownership, pride, ease, expression, helpfulness, collaboration, uniqueness, enjoyment, and surprise. Free-form responses were also collected, and quantitative results were compared descriptively with Dramatron.

Limitations

  • The paper states that documentary filmmakers’ emphasis on factual fidelity was outside the study’s focus and likely contributed to their lower ownership and pride scores. It also notes that AI responsiveness depends on GPT/OpenAI API response times, that prototype bugs affected usability, that GPT-4 tends toward safer and more conventional ideas, and that UI and character-development features need improvement.
  • Evident from the text: the sample is small (23) and unevenly split across subgroups (9 screenwriters, 9 documentary filmmakers, 5 VFX artists), the task was short (about 30 minutes), and the main outcomes are self-reported Likert ratings. The comparison with Dramatron is cross-study rather than a controlled head-to-head test.

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

Integrity signals

Registry

Crossref · 10.1145/3639701.3656325

Cites

17 of 17 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

14 of 26 cited sources checked · 6 supported · 1 flagged · accuracy not yet measured

  • Interpretationnone given

    Pride in the outcome was lower for wr-AI-ter (46%) than for Dramatron.

    “However, pride of outcome was lower for wr-AI-ter (46%) compared to Dramatron.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.4 assigns the 46% pride figure to participants evaluating Dramatron, not wr-AI-ter. The paper does not make the claimed comparison between the two tools. (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
  • Premisecitation

    More than half of Dramatron's respondents did not feel ownership of AI-generated scripts.

    “After realizing that more than half of Dramatron's respondents did not feel ownership of the scripts generated by the AI”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source
  • Interpretationown data

    Ease-of-use agreement was lower for wr-AI-ter (48%) than for Dramatron (61%).

    “Ease of use was a split opinion for Dramatron (61% agreement), while wr-AI-ter experienced a lower agreement (48%).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.4 reports that opinion on Dramatron’s ease of use was split, with 61% agreeing. The cited paper’s survey concerns Dramatron; it does not report the claimed 48% agreement for wr-AI-ter or establish the comparison. (LLM, conf=0.88)

      Experimental — accuracy not yet measured
  • Study design and conductreasoning

    The surprise question (Q9) is no longer comparable with the Dramatron survey because LLMs were less well known at that time.

    “Therefore, Q9 is no longer comparable with the answers of the participants in the Dramatron survey, as LLMs were even less well known at that time.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1 confirms that the Dramatron survey asked whether participants were surprised by the AI system’s responses, and §6.1.6 raises novelty as a possible confounding factor. The paper does not say that the citing paper’s Q9 is no longer comparable with its survey or that LLMs were less well known at the time; those parts of the claim are unsupported by our copy. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
  • Findingown data

    Screenwriters showed 78% agreement on ownership.

    “followed closely by screenwriters at 78% (44% agreed and 33% strongly agreed).”
    • Also stated in: § 7 DISCUSSION 7.1 Insights from the Quantitative Answers and Comparison to Dramatron · plain
      “Similar results were found for the screenwriters (78% agreed, none disagreed).”
    Quote in the paperSubstantial claim
  • May combine two claims. 27
  • The quote only partly states this claim. 3
  • Classified as probably not a substantial claim. 36

  • May combine two claims.

    Interpretationown data

    Helpfulness approval was 71% for wr-AI-ter and 84% for Dramatron.

    “Both tools were seen as helpful, with Dramatron receiving an 84% approval rating in this regard, compared to wr-AI-ter's substantial 71%.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.2 reports that 84% of respondents agreed Dramatron was helpful. The cited paper's survey results do not provide the claimed 71% helpfulness rating for wr-AI-ter, so only the Dramatron part is supported. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • The quote only partly states this claim.

    Interpretationown data

    wr-AI-ter produced a higher sense of ownership (57%) than Dramatron.

    “Our results showed a higher sense of ownership (57%) for the results generated by wr-AI-ter”
    • Also stated in: § 7 DISCUSSION 7.1 Insights from the Quantitative Answers and Comparison to Dramatron · plain
      “when compared to Dramatron, our app increased the perception of ownership.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.5 reports low ownership of scripts co-written with Dramatron: more than half of respondents felt they did not own the result. The paper does not evaluate wr-AI-ter or report its 57% figure, which the citing sentence presents as the citing authors’ own result. It supports the Dramatron side of the comparison, but not the full comparative claim. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationcitation

    Collaboration agreement was 43% for wr-AI-ter and 77% for Dramatron.

    “the sense of collaboration with the AI system was more pronounced in Dramatron's evaluation (77% agreement) compared to our fndings (43% agreement)”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

  • May combine two claims.

    Interpretationown data

    Uniqueness was reported by 14% of wr-AI-ter users and 69% of Dramatron users.

    “The perception of uniqueness presents a signifcant divergence, with Dramatron's outputs being considered unique by 69% of participants, in contrast to wr-AI-ter's 14%.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.1 reports that 69% of Dramatron survey respondents agreed the scripts felt unique. Our copy does not report a 14% uniqueness rating for wr-AI-ter or a comparison between the two tools (searched: “wr-AI-ter 14% uniqueness” and “unique 69% 14% comparison”). No page number was available in the indexed passages. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationown data

    Expression of creative goals was affirmed by 21% of wr-AI-ter users, compared with 77% of Dramatron users.

    “Dramatron's capacity for expressing creative goals was afrmed by 77% of its users, a contrast to the 21% in our study.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

  • May combine two claims.

    Interpretationown data

    Enjoyment was 71% for wr-AI-ter and 77% for Dramatron.

    “The enjoyment levels (71% for wr-AI-ter) remain high, albeit slightly lower than Dramatron's 77%”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.1 reports that 77% of Dramatron survey respondents agreed they enjoyed writing with the AI system. The paper does not report a 71% enjoyment figure for wr-AI-ter, so it supports only the Dramatron side of the comparison. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationcitation

    The wr-AI-ter results compare favorably to Dramatron but are slightly less positive than Wordcraft.

    “Additionally, our results compare favorably to those from Dramatron but are slightly less positive when compared to Wordcraft.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source
    • cites [35] 35. Writesonic. 2023. AI Content Creation & Customer Experience Platform. Retrieved February 7, 2024 from https://writesonic.com/

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductcitation

    The study used 5-point Likert-type scales and adapted the same questions as Dramatron to ensure comparability.

    “To ensure optimal comparability with the Dramatron framework, our methodology adapted the same questions highlighted in [25,35], employing 5-point Likert-type scales.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source
    • cites [35] 35. Writesonic. 2023. AI Content Creation & Customer Experience Platform. Retrieved February 7, 2024 from https://writesonic.com/

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Premisecitation

    Dramatron is seen not as a professional scriptwriting tool but as a learning tool or resource.

    “One interpretation was that Dramatron is not seen as a professional scriptwriting tool, but rather as a learning tool to practice writing or as a resource [25].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

  • May combine two claims.

    Interpretationcitation

    Surprise was 35% for wr-AI-ter and 92% for Dramatron.

    “The diference in the element of surprise between wr-AI-ter and Dramatron (35% vs. 92%)”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [25] 25. Piotr Mirowski, Kory W. Mathewson, Jaylen Pittman, and Richard Evans. 2023. Co-Writing Screenplays and Theatre Scripts with Language Models: Evaluation by Industry Professionals. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 355, 34 pages. https://doi. org/10.1145/3544548.3581225 source

      Section 6.1.1 reports that 92% of Dramatron survey respondents agreed they were surprised by the AI system’s responses. Our copy does not report a 35% surprise figure for wr-AI-ter or compare the two systems on that measure (searched: “wr-AI-ter” and “35%”). No page number was available for the passage. (LLM, conf=0.87)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the wr-AI-ter study, 21% of participants did not find the results surprising.

    “Meanwhile, a dissenting 21% did not fnd the results surprising (14% disagreed and 7% strongly disagreed)”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    In the wr-AI-ter study, 43% of participants recognized collaboration with the AI.

    “In terms of collaboration, 43% of participants recognized its value (14% agreed and 29% strongly agreed).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Users expressed a recurring desire for the AI to be more responsive to input and to allow deeper exploration of ideas.

    “There was a recurring desire for the AI to be more responsive to user input and to allow deeper exploration of ideas (P2, P3, P5).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Among scriptwriters, 33% disagreed about pride, and none strongly disagreed.

    “in contrast to the group of scriptwriters where 33% disagreed, albeit without any instances of strong disagreement”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    There was a critique that the stories were generic and predictable (P19).

    “There was a critique about the generic and predictable nature of the stories (P19)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Documentary filmmakers showed 78% lack of ownership, with 67% disagreeing and 11% strongly disagreeing.

    “Conversely, a substantial majority of documentary flmmakers, 78%, reported a lack of ownership (67% disagreed and 11% strongly disagreed).”
    • Also stated in: § 7 DISCUSSION 7.1 Insights from the Quantitative Answers and Comparison to Dramatron · boosted
      “Documentary flmmakers, on the other hand, disagreed with a clear majority of 78%, 22% agreed.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the wr-AI-ter study, 74% of respondents provided open-ended free-form responses.

    “74% of respondents ofered concise, open-ended responses.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Findingnone given

    During preliminary evaluation, if brainstorming lacked depth and attention to detail, the output story became more generic and described an abstract story structure.

    “If the brainstorming lacked depth and attention to detail, the output story became more generic and described an abstract story structure.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    A participant (P8) mentioned the need for better character development.

    “A participant (P8) mentioned the need for better character development”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the wr-AI-ter study, 14% of participants did not agree that they enjoyed the interaction.

    “However, there was a small contingent of 14% who did not agree and didn't share a positive experience.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In the wr-AI-ter study, 29% disagreed on expression, and none strongly disagreed.

    “while 29% were in disagreement, and notably, no participants expressed strong disagreement.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    No participant strongly disagreed about enjoyment.

    “No participant strongly disagreed with Q8.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    No respondents strongly disagreed about ease of use.

    “It is notable that none of the respondents strongly disagreed with the ease of use”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Some users commented that the AI integration was intuitive but that the interface could be improved for beginners.

    “Some users (P6, P7) commented on the intuitive nature of the AI's integration and guided software experience, but also noted that the interface could be improved for better user experience, especially for beginners”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    There was an acknowledgment (P23) that human intervention is still necessary to refine and finalize scripts.

    “There was an acknowledgment (P23) that while the app helps to create content, human intervention is still necessary to refne and fnalize the scripts.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the wr-AI-ter study, 14% of participants agreed that the results were unique.

    “Merely 14% of respondents agreed that the results produced by the application were unique.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Documentary filmmakers showed 11% agreement and 44% disagreement on ease of use.

    “Documentary flmmakers, however, showed minimal agreement at only 11%, with 44% expressing disagreement with Q3.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the wr-AI-ter study, 35% of participants were surprised by the tool's outcomes.

    “Only 35% of the study participants were surprised by the outcomes generated by the tool (21% agreed and 14% strongly agreed).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S1 (asserted), S2 (checkable), S3 (load-bearing).

    Findingown data

    The interface could be improved for beginners who are unsure where to start.

    “but also noted that the interface could be improved for better user experience, especially for beginners who are not sure about where to start.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S1 (asserted), S2 (checkable), S3 (load-bearing).

    Findingown data

    There is a need for more sophisticated features for character development.

    “A participant (P8) mentioned the need for better character development, indicating a desire for more sophisticated features that can handle complex elements of storytelling.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Participants desired more control over the generation of their ideas.

    “Users desire more control over the generation of their ideas.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingnone given

    User feedback highlighted the tool's potential but also suggested areas for improvement.

    “User feedback highlighted the tool's potential but also suggested areas for improvement.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The app was not as useful for finalizing scripts.

    “Though it was not as useful for fnalizing scripts.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    17% of participants reported no previous experience with AI writing tools, while 83% had prior experience.

    “17% of the participants reported no previous experience with AI writing tools, whereas a substantial 83% had prior experience utilizing such technologies.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationown data

    The AI's ability to transform chaotic brainstorming into a structured narrative with a consistent plot can help screenwriters overcome challenges like writer's block or ordering ideas.

    “The AI's ability to transform even the most chaotic brainstorming session into a structured narrative with a consistent plot can help screenwriters overcome common challenges, like writer's block or ordering many ideas into a useful and smooth structure.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Film and series production is typically highly collaborative, often involving writers' rooms.

    “Film and series production is typically a highly collaborative process.Creative teams often come together in so-called”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [17] 17. Miira Karhula, Timo Lehti, and Teppo Nuutinen. 2021. The Writers' Room. (2021).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Existing AI-screenwriting systems often lack context awareness, interpreting only selected text and not the holistic story arc.

    “Another constraint revolves around the system's lack of context awareness.The system often interprets only the selected text section and does not consider the holistic story arc [32].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [32] 32. Weber C. J. et al.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    If successive scenes fail to raise questions, audience interest may diminish; unresolved questions may lead to dissatisfaction.

    “If successive scenes fail to raise questions, audience interest may diminish.Conversely, unresolved questions may lead to dissatisfaction.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Pleasure-Arousal-Dominance Model highlights certain emotional combinations that enhance the likelihood of a video gaining widespread attention.

    “The Pleasure-Arousal-Dominance Model [24] highlights certain emotional combinations that enhance the likelihood of a video gaining widespread attention.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [24] 24. Albert Mehrabian. 1996. Pleasure-arousal-dominance: A general framework for describing and measuring individual diferences in Temperament. Current Psychology 14, 4 (01 Dec 1996), 261-292. https://doi.org/10.1007/BF02686918 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Research suggests that videos often go viral when the right combinations of emotions are presented.

    “Research suggests that videos often go viral when the right combinations of emotions are presented [18].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [18] 18. Jones Kerry, Libert Kelsey, and Tynski Kristin. 2016. The Emotional Combinations That Make Stories Go Viral. Retrieved February 7, 2024 from https://hbr.org/ 2016/05/research-the-link-between-feeling-in-control-and-viral-content

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    According to [23], a successful story beat typically resolves a question while introducing a new one, and this pattern is essential for evaluating story tension.

    “According to [23], a successful story beat typically resolves a question while introducing a new one.This pattern of raised and resolved questions is essential for evaluating story tension.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [23] 23. Robert McKee. 2010. Story: Style, Structure, Substance, and the Principles of Screen- writing. HarperCollins.

      Not checked in the free run.

  • May combine two claims.

    Premisenone given

    Existing AI-screenwriting tools take little user input to guide generation and explore interaction and co-creation minimally.

    “Furthermore, the tools may suggest rewriting a particular passage, but they take very little input from the user to guide the generation process.The ways of interactions and co-creation with AI are explored only minimally.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Wordcraft users reported high ownership over AI co-created works, but Wordcraft was designed solely for short stories and not full scripts, and its stories often lacked contextual awareness.

    “Although users reported a high level of ownership over the works they cocreated with AI, it is important to note that Wordcraft was designed solely for short stories and not full scripts and the stories often lacked contextual awareness.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [35] 35. Writesonic. 2023. AI Content Creation & Customer Experience Platform. Retrieved February 7, 2024 from https://writesonic.com/

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationreasoning

    The design principle of not allowing direct insertion of AI suggestions safeguards the writer's ultimate control over the narrative and prevents undue authorship by the AI.

    “This design principle safeguards the writer's ultimate control over the narrative, preventing undue authorship by the AI.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Study design and conductnone given

    The AI's responsiveness is limited by GPT's response time through the OpenAI API.

    “One limitation worth noting is the AI's responsiveness, which is dependent on GPT's response time through the OpenAI API.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The format and scope of creative tasks can influence a user's experience with AI tools.

    “This could mean that the format and scope of creative tasks can infuence the user's experience with AI tools.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The app's value extends beyond content generation to sparking new thoughts and facilitating creative brainstorming.

    “This positive reception underscores the tool's potential as a catalyst for creativity, suggesting that its value extends beyond mere content generation to include sparking new thoughts and facilitating the creative brainstorming process.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The tool promotes users' creative skills but does not replace them.

    “The tool promotes the creative skills of its users but does not replace them.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    A lower surprise score could be due to a higher ownership score.

    “Accordingly, a lower value in surprise could be due to the higher value in ownership.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The app's role is in the early stages of creative work.

    “This perception reinforces the app's role in the early stages of creative work”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Users appreciate the app's ability to provide creative inspiration and its ease of use.

    “The results of our study implicate that users appreciate the app's ability to provide creative inspiration and its ease of use.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationown data

    Feedback about generic and predictable outputs highlights a challenge in content diversity and originality.

    “Feedback about the generic and predictable nature of some outputs highlights a challenge in content diversity and originality.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims. The quote only partly states this claim.

    Interpretationreasoning

    Bugs impaired user-friendliness and dampened enjoyment of the tool.

    “The impairment caused by bugs can dampen the pleasure of trying out the tool.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Wordcraft project showed that users can experience a strong sense of ownership when creating short stories.

    “The Wordcraft project showed that users can experience a strong sense of ownership when creating short stories [35].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [35] 35. Writesonic. 2023. AI Content Creation & Customer Experience Platform. Retrieved February 7, 2024 from https://writesonic.com/

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Interpretationown data

    The app needs to produce more varied and less formulaic content.

    “which suggests a need for the app to produce more varied and less formulaic content.”
    Quote in the paperNot a substantial claim
55 of 55

  • The paper relies on this without support.

    When AI is used in screenwriting, the story's meaning must not be overshadowed for users to feel ownership.

    “When leveraging AI in the screenwriting process, it is essential for the feeling of ownership that the meaning is not overshadowed.”

  • The paper relies on this without support.

    Documentary filmmakers attach great importance to factual fidelity.

    “An exception were the documentary flmmakers, who attach great importance to factual fdelity, which was not the focus of this paper.”

  • The paper relies on this without support.

    GPT-4 tends to produce safer, more conventional ideas.

    “GPT-4 is designed to produce results that tend toward safer, more conventional ideas.”
    “Furthermore, GPT-4 tends to produce output that gravitates towards safer, more conventional ideas.”

  • The paper relies on this without support.

    Users are already familiar with AI-generated content and are therefore less surprised by default.

    “While it is true that users are already familiar with AI-generated content and are therefore less surprised by default”

  • The paper relies on this without support.

    In scriptwriting, ownership and authorship are closely related, and the legal aspects surrounding these issues are currently unresolved.

    “In scriptwriting, the concepts of ownership and authorship are closely related, and the legal aspects surrounding these issues are currently unresolved.”
5 of 5

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Documentary filmmakers' pride agreement changes from 44% to 11%

    Major

    The results say 44% of documentary filmmakers felt proud, whereas the discussion says only 11% agreed. The latter matches the earlier strongly-agreed share, not total agreement.

  • Overall ease-of-use agreement conflicts with subgroup results

    Major

    The reported subgroup rates imply 5 agreeing screenwriters, 3 VFX artists, and 1 documentary filmmaker. That is 9 of 23 participants, or about 39%, rather than the reported 48%.

  • Overall pride agreement changes from 39% to 46%

    Major

    The results give 39% as the share proud of the outputs, but the discussion uses 46% for that same outcome.

  • Generic writing-tool citation points to a transformer paper

    Minor

    Reference [34] is cited as an example of a generic text tool, but its bibliography entry is the paper "Attention is All you Need."

  • VFX artists' pride agreement changes from 80% to 60%

    Minor

    The results report that 80% of VFX artists agreed or strongly agreed that they were proud of the outputs; the discussion describes their agreement as 60%.

  • Wordcraft citation points to Writesonic

    Minor

    The text attributes a study by Yuan and colleagues to reference [35], but [35] is a Writesonic entry. The Yuan and colleagues Wordcraft paper appears as reference [36].

6 of 6

Cites

Flagged first, then by citations

Research works printed but not linked: 8

  1. Jasper.AI. 2023. Jasper AI Copilot. Retrieved February 7, 2024 from https://www.jasper.ai/
  2. Miira Karhula, Timo Lehti, and Teppo Nuutinen. 2021. The Writers’ Room. (2021).
  3. Stephen King. 2000. On writing: A memoir of the craft. Simon and Schuster.
  4. Misha Laskin, Kimin Lee, Adam Stooke, Lerrel Pinto, Pieter Abbeel, and Aravind Srinivas. 2020. Reinforcement learning with augmented data. Advances in neural information processing systems 33 (2020), 19884–19895.
  5. Robert McKee. 2010. Story: Style, Structure, Substance, and the Principles of Screenwriting. HarperCollins.
  6. ShortlyAI. 2021. Make writer’s block a thing of the past. Retrieved February 7, 2024 from https://www.shortlyai.com/
  7. Blake Snyder. 2023. SAVE THE CAT!®. Noura Books.
  8. Writesonic. 2023. AI Content Creation & Customer Experience Platform. Retrieved February 7, 2024 from https://writesonic.com/

Cited by

Sorted by citations

Related

Authors
Weber, C. J., Burgkart, S., & Rothe, S.
Affiliations
  • University of Television and Film Munich (DE)
  • Ludwig-Maximilians-Universität München (DE)
ORCID
Venue
ACM International Conference on Interactive Media Experiences
Publisher
ACM
Published
2024-06-07
Pages
145-156
DOI
10.1145/3639701.3656325
OpenAlex
W4399428753
Semantic Scholar
6b5e4351e7641280839ecaf0e7e7d73810b15cef
Cited by
13
References
17
Citations by year
2026: 3 · 2025: 8 · 2024: 2
FWCI
70.57
Type
conference paper
Language
en
Open access
Licence
www.acm.org/publications/policies/copyright_policy#Background (vor)
Locations
ACM International Conference on Interactive Media Experiences (publishedVersion)
Topics
  • Social Sciences › Arts and Humanities › Visual Arts and Performing Arts › Artistic and Creative Research
  • Physical Sciences › Computer Science › Human-Computer Interaction › Virtual Reality Applications and Impacts
Keywords
authorship, creative process, interactive applications, idea generation, script structure, user study, documentary filmmaking

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.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-09-30.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-09-30.

Cited sources are checked by 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