請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/102786| 標題: | Book WEDO 商業計畫書 AI時代的數位創作者平台 —以行動研究法探討AI協作之Solo創業實踐 Book WEDO: A Digital Creator Platform for the AI Era An Action Research Study of Solo Entrepreneurship through Human-AI Collaboration |
| 作者: | 洪儒明 Ju-Ming Hung |
| 指導教授: | 李家岩 Chia-Yen Lee |
| 關鍵字: | AI協作創業; 行動研究; Solo創業; 知識資產平台; 創作者經濟; 平台治理; Book WEDO AI-assisted entrepreneurship; action research; solo entrepreneurship; knowledge-asset platform; creator economy; platform governance; Book WEDO |
| 出版年 : | 2026 |
| 學位: | 碩士 |
| 摘要: | 生成式AI正在降低內容產製與軟體開發的門檻,也使個人創業者得以用更小的組織完成過去需要多職能團隊分工的工作。這項變化不只關乎效率提升,也重新提出一個更具實務性的問題:AI協作能否支撐長篇知識內容的結構化、權限治理、讀者互動與商業化。本研究以Book WEDO為行動研究個案,探討一位Solo創業者如何從自身出版需求出發,運用AI協作將單書發布工具逐步轉化為AI-ready知識資產與IP孵化基礎設施。
本研究採行動研究法,資料蒐集期間為2025年12月16日至2026年5月31日。研究資料包含開發日誌、Git版本歷程、AI對話紀錄、平台實作成果、出版與市場早期訊號、外部案例觀察,以及六組共38位質化有效樣本。研究過程依三輪循環推進:第一輪建立單書MVP與長篇閱讀體驗;第二輪擴展多作者流程、權限治理與 AI 伴讀;第三輪透過訪談資料與外部案例校正平台定位、商業模式與後續驗證指標。上述資料形式的證據強度並不相同,因此本文以多元資料來源交叉佐證作為主要推論原則,而不將38位樣本視為大規模量化市場代表樣本。 研究發現,AI協作能明顯縮短產品原型與內容整理週期,但無法取代創業者在問題定義、功能取捨、風險控管與品質驗收上的責任。Book WEDO在研究過程中由「長篇數位書籍發布工具」修正為「AI-ready知識資產與IP孵化基礎設施」;其商業模式也由單一工具訂閱,調整為知識解鎖、工具訂閱、知識採礦服務、IP孵化與企業合作並行的五層收入模型。Book WEDO 2.0已完成產品基線與定位校正,但尚不能等同於產品市場契合,仍需以作者留存、讀者付費、企業PoC與AI伴讀可信度等指標持續驗證。 本研究提出「船長架構」作為AI協作Solo創業的決策分工框架,說明AI可以放大個人創業者的產品建構能力,但創業優勢仍取決於人是否能將AI產出納入可追溯、可驗證、可治理的產品與商業流程。本文的貢獻不在於宣稱一人即可取代組織,而在於呈現當AI降低執行成本後,創業者更需要被重新訓練的能力:判斷、取捨、治理與長期承擔。後續研究可進一步比較不同產業的AI協作創業案例,並追蹤知識資產平台在付費、企業導入與內容治理上的長期實證資料。 Generative AI is lowering the barriers to content production and software development, allowing solo entrepreneurs to perform work that previously required multi-functional teams. This shift raises a practical question beyond efficiency: whether AI collaboration can support the structuring, governance, reader interaction, and commercialization of long-form knowledge. This study examines Book WEDO as an action research case and investigates how a solo entrepreneur transformed a first-hand publishing need into an AI-ready knowledge-asset and IP incubation infrastructure. The study adopts action research, with data collected from December 16, 2025 to May 31, 2026. The empirical materials include development logs, Git version histories, AI conversation records, platform implementation evidence, early publishing and market signals, external case observations, and 38 valid qualitative samples across six stakeholder groups. The research proceeded through three action cycles: building a single-book MVP and long-form reading experience; extending the platform toward multi-author workflows, permission governance, and AI-assisted reading; and refining the platform positioning, business model, and validation indicators through interview data and external case comparison. Because these materials differ in evidentiary strength, the study relies on multi-source cross-verification rather than treating the 38 samples as a large-scale quantitative market sample. The findings show that AI collaboration can substantially shorten the cycle of product prototyping and content organization, but it does not replace the entrepreneur\'s responsibility for problem definition, product prioritization, risk control, and quality assurance. During the research process, Book WEDO evolved from a long-form digital publishing tool into an AI-ready knowledge-asset and IP incubation infrastructure. Its business model also shifted from a single subscription logic to a five-layer revenue framework consisting of knowledge unlocking, tool subscription, knowledge mining services, IP incubation, and enterprise collaboration. The completion of Book WEDO 2.0 represents a product baseline and positioning validation, not product-market fit; further evidence is still required from author retention, reader payment behavior, enterprise PoC outcomes, and trustworthy AI-assisted reading. This study proposes the \"Captain Framework\" as a decision-allocation model for AI-assisted solo entrepreneurship. The Book WEDO case shows that AI can expand the product-building capacity of an individual entrepreneur, while the core entrepreneurial advantage still depends on whether AI outputs can be incorporated into traceable, verifiable, and governable product and business processes. The contribution of this study is not to claim that one person can replace an organization, but to show that when AI reduces execution costs, entrepreneurs must strengthen judgment, prioritization, governance, and long-term accountability. Future research may compare AI-assisted solo entrepreneurship across industries and examine long-term evidence of payment behavior, enterprise adoption, and knowledge governance in creator knowledge-asset platforms. |
| URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/102786 |
| DOI: | 10.6342/NTU202601279 |
| 全文授權: | 同意授權(全球公開) |
| 電子全文公開日期: | 2026-07-23 |
| 顯示於系所單位: | 創業創新管理碩士在職專班(EiMBA) |
文件中的檔案:
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-114-2.pdf | 4.39 MB | Adobe PDF | 檢視/開啟 |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。
