[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-1-en-105":3,"doc-seo-571412-105":53,"doc-detail-571412-en":120},{"code":4,"msg":5,"data":6},0,"success",[7,14,19,24,29,34,39,44,49],{"id":8,"doc_module":9,"doc_module_name":10,"category_name":11,"show_sort_weight":12,"slug":13},11,1,"Template","Presentations",90,"presentations",{"id":15,"doc_module":9,"doc_module_name":10,"category_name":16,"show_sort_weight":17,"slug":18},12,"Resumes",80,"resumes",{"id":20,"doc_module":9,"doc_module_name":10,"category_name":21,"show_sort_weight":22,"slug":23},14,"Invoices",70,"invoices",{"id":25,"doc_module":9,"doc_module_name":10,"category_name":26,"show_sort_weight":27,"slug":28},15,"Posters",60,"posters",{"id":30,"doc_module":9,"doc_module_name":10,"category_name":31,"show_sort_weight":32,"slug":33},16,"Social Media",50,"social-media",{"id":35,"doc_module":9,"doc_module_name":10,"category_name":36,"show_sort_weight":37,"slug":38},17,"Forms",40,"forms",{"id":40,"doc_module":9,"doc_module_name":10,"category_name":41,"show_sort_weight":42,"slug":43},18,"Letters",30,"letters",{"id":45,"doc_module":9,"doc_module_name":10,"category_name":46,"show_sort_weight":47,"slug":48},21,"Paper Templates",5,"papers-templates",{"id":50,"doc_module":9,"doc_module_name":10,"category_name":51,"show_sort_weight":4,"slug":52},158,"General","general-158",{"code":4,"msg":54,"data":55},"ok",{"site_id":56,"language":57,"slug":58,"title":59,"keywords":60,"description":61,"schema_data":62,"social_meta":113,"head_meta":115,"extra_data":117,"updated_unix":119},105,"en","ai-driven-game-design-game-design-document-generation-using-large-language-models-and-langgraph-bachelor-thesis","AI-Driven Game Design - Game Design Document Generation Using Large Language Models and LangGraph - Bachelor Thesis","","AI-Driven Game Design investigates how effectively large language models can support game design by generating ideas and planning a Game Design Document (GDD). The work minimizes human input to evaluate whether an AI agentic workflow can produce a fully developed, structurally sound GDD with minimal intervention. Three research questions assess originality, internal logic and technical consistency, and the communicative quality and readability for human understanding. Results show more innovative concept generation, but weaker structural integrity and clarity, with fluff-heavy mechanics and limited divergent thinking.",{"@graph":63,"@context":112},[64,80,95],{"@type":65,"itemListElement":66},"BreadcrumbList",[67,71,74,77],{"item":68,"name":69,"@type":70,"position":9},"https://docshare.wps.com","Home","ListItem",{"item":72,"name":10,"@type":70,"position":73},"https://docshare.wps.com/template/",2,{"item":75,"name":46,"@type":70,"position":76},"https://docshare.wps.com/template/paper-templates/",3,{"item":78,"name":59,"@type":70,"position":79},"https://docshare.wps.com/template/ai-driven-game-design-game-design-document-generation-using-large-language-models-and-langgraph-bachelor-thesis/571412/",4,{"url":78,"name":59,"@type":81,"author":82,"headline":59,"publisher":85,"fileFormat":88,"inLanguage":57,"description":61,"dateModified":89,"datePublished":89,"encodingFormat":88,"isAccessibleForFree":90,"interactionStatistic":91},"DigitalDocument",{"name":83,"@type":84},"Riley West","Person",{"url":68,"name":86,"@type":87},"DocShare","Organization","application/pdf","2026-10-13",true,{"@type":92,"interactionType":93,"userInteractionCount":4},"InteractionCounter",{"@type":94},"ViewAction",{"@type":96,"mainEntity":97},"FAQPage",[98,104,108],{"name":99,"@type":100,"acceptedAnswer":101},"How does the thesis evaluate LLMs for game design and GDD creation?","Question",{"text":102,"@type":103},"It generates game ideas and plans a GDD with minimal human input, focusing on how well the AI can produce a complete and usable document. Three research questions measure different GDD qualities.","Answer",{"name":105,"@type":100,"acceptedAnswer":106},"What are the main strengths and limitations found in the AI-generated GDDs?",{"text":107,"@type":103},"The AI process creates more innovative and creative game ideas than humans. However, the documents still lack structural integrity and clarity, often using mechanics keywords without sufficient context.",{"name":109,"@type":100,"acceptedAnswer":110},"Why is full automation considered insufficient for high-level creativity in this work?",{"text":111,"@type":103},"The findings suggest automation lacks the divergent thinking needed for high-level creativity. Repeated runs with similar inputs tend to converge toward similar outputs, limiting novelty.","https://schema.org",{"og:url":78,"og:type":114,"og:title":59,"og:site_name":86,"og:description":61},"article",{"robots":116,"canonical":78},"index,follow",{"doc_id":118,"site_id":56},571412,1791916460,{"code":4,"msg":5,"data":121},{"doc_id":118,"user_id":122,"nickname":83,"user_avatar":123,"doc_module":9,"category_id":45,"category_name":46,"doc_title":59,"doc_description":61,"doc_content":124,"file_id":125,"file_url":126,"file_type":127,"file_size":128,"view_count":4,"is_deleted":4,"is_public":9,"is_downloadable":9,"audit_status":9,"page_count":129,"language":130,"language_code":57,"site_id":56,"html_lang":57,"table_of_contents":131,"faqs":132,"seo_title":133,"seo_description":61,"update_tm":119,"read_time":25},1099523885074,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","BACHELOR’S THESIS  \nAI-Driven Game Design  \n„Game Design Document Generation Using Large Language  \nModels and LangGraph“  \nverfasst von / submitted by  \nJonathan Wurth  \nangestrebter akademischer Grad / in partial fulfilment of the requirements for the degree of  \nBachelor of Science (BSc)  \nWien, 2026 / Vienna, 2026  \nStudienkennzahl lt. Studienblatt / UA 033 521  \ndegree programme code as it appears on the student record sheet:  \nStudienrichtung lt. Studienblatt / Bachelor Informatik  \ndegree programme as it appears on the student record sheet:  \nBetreut von / Supervisor: Univ. Prof. Dipl.-Ing. Dr. Helmut Hlavacs  \nMitbetreut von / Co-Supervisor:  \nContents  \n1. Introduction 1  \n2. Related Work 3  \n3. Technical Foundation 9  \n3.1. Game Design Document ............................ 9  \n3.2. LangChain ................................... 10  \n3.3. LangGraph ................................... 12  \n3.4. Gemini ...................................... 13  \n4. Creating Game Design Documents with LLMs 15  \n4.1. Prompt Engineering .............................. 16  \n4.2. Architectual worklow .............................. 17  \n4.3. Agent Architecture ............................... 18  \n4.4. GDDState Structure .............................. 21  \n4.5. Phased Questioning and GDD Templates .................. 22  \n5. Evaluation 23  \n5.1. Experimental Setup and Methodology .................... 23  \n5.2. Evaluation Metrics and Questionnaire Design ................ 24  \n5.3. Statistical Analysis and Inference ....................... 25  \n5.4. Discussion and Qualitative Findings ..................... 26  \n6. Conclusion 29  \nBibliography 31  \nA. Appendix 35  \n1 . Introduction  \nGaming is, with good reason, the largest and highest-grossing entertainment sector in the world. The fascinating properties of games can help humans learn and safely gain experiences without the fear of real-world consequences. Through gameplay, humans can explore emotions and experiment with different strategies, which can help them master the real world. A Game Design Document (GDD) is used by the game designer to communicate with the development team to ensure everyone is on the same page. A GDDis created at the start of a game’s lifecycle and iterated upon until the game is released. It contains not only the theoretical mechanics and gameplay loops of the game but also handles how the game should look and feel, as well as the technical engine, database, and programming language the game will be developed in. Because of this, it is necessary to have a GDD as early as possible to communicate the game designer’s needs to the team.  \nComing up with a Game Design Document is not an easy task. It is essentially conceiving a game idea, fleshing it out, and figuring out all the intricate details of how you want your game to look, feel, and what experience you want the players to have. This is a formidable task, especially for newer game developers. Because of this time-intensive and complex process, people seek ways to simplify it. With the rise of AI and Large Language Models (LLMs), new possibilities have been created to streamline the ideation and GDD writing process. As in most creative activities, LLMs can be used as a tool to help create games and make them more immersive. For example, in-game, LLMs can be used for adaptive storylines and quest structures in which NPCs can react with more freedom to the player’s actions. They find even more utility in the production phase of video game creation. LLMs can be used for repetitive tasks, like coding with Copilot, helping in writing storylines for the world and characters, and creating concept art and in-game assets with generative models like Midjourney. In almost every step, you can enhance your performance with LLMs.  \nThis thesis seeks to investigate how effectively LLMs can be used for game design by creating ideas and planning a GDD for video games. Therefore, as little human input as possible is given to the LLMs to","cbCailYfRejVP5qD","https://ap.wps.com/l/cbCailYfRejVP5qD","pdf",660871,43,"English","# Introduction\n## Research questions\n# Related Work\n# Technical Foundation\n## Game Design Document\n## LangChain\n## LangGraph\n## Gemini\n# Creating Game Design Documents with LLMs\n## Prompt Engineering\n## Agent Architecture\n## GDDState Structure\n## Phased Questioning and GDD Templates\n# Evaluation\n## Experimental Setup and Methodology\n## Evaluation Metrics and Questionnaire Design\n## Statistical Analysis and Inference\n## Discussion and Qualitative Findings\n# Conclusion\n# Bibliography\n# Appendix","[{\"question\":\"How does the thesis evaluate LLMs for game design and GDD creation?\",\"answer\":\"It generates game ideas and plans a GDD with minimal human input, focusing on how well the AI can produce a complete and usable document. Three research questions measure different GDD qualities.\"},{\"question\":\"What are the main strengths and limitations found in the AI-generated GDDs?\",\"answer\":\"The AI process creates more innovative and creative game ideas than humans. However, the documents still lack structural integrity and clarity, often using mechanics keywords without sufficient context.\"},{\"question\":\"Why is full automation considered insufficient for high-level creativity in this work?\",\"answer\":\"The findings suggest automation lacks the divergent thinking needed for high-level creativity. Repeated runs with similar inputs tend to converge toward similar outputs, limiting novelty.\"}]","AI-Driven Game Design - Game Design Document Generation Using Large Language Models and LangGraph - Bachelor Thesis | PDF"]