本研究探討生成式人工智慧（Generative AI）在未經授權資料使用的情境下，政府採行罰款制度對市場結構與參與行為的影響，並進一步分析其對社會福利的效果。本文建構一賽局模型，納入政府、AI公司、創作者與消費者四方，評估在不同罰款強度下，各方的進場誘因與市場均衡。研究結果指出，政府最適政策應視市場組合而定。當創作者即使在未對AI公司處以罰款的情況下也願意進場時，貿然課罰可能破壞市場競爭，反而降低社會福利。相反地，若創作者在沒有罰款的情況下缺乏進場誘因，則提高罰款以迫使AI公司退出，能為創作者進場創造空間，並提升市場效率與整體福利。本文強調，制度設計的關鍵不在於罰款的高低，而在於是否能創造雙方都有誘因參與市場的穩定均衡。若僅仰賴懲罰手段，可能無法同時促使兩方進場。因此，建議政府搭配授權機制或補償制度等其他工具，以確保政策效果與初衷一致。

This thesis analyzes how government fines affect market structure when Generative AI is trained on unlicensed data. A game-theoretic model is developed featuring four key actors—government, AI firms, creators, and consumers—to analyze entry incentives and equilibrium outcomes under varying levels of fines. The results suggest that the optimal policy depends on the prevailing market structure. If creators are willing to enter the market even without penalties on Generative AI firms, imposing such fines may disrupt competition and reduce social welfare. In contrast, if creators lack the incentive to enter under zero-penalty conditions, increasing fines can drive AI firms out, create space for creators to participate, and enhance both market efficiency and social welfare. The study highlights that the core of institutional design lies not in the severity of fines, but in whether the policy facilitates a stable equilibrium where both sides are willing to enter the market. Relying solely on punishment may fail to achieve mutual participation and risk undermining policy goals. Therefore, the government should consider additional tools—such as licensing mechanisms or compensation schemes—to ensure that policy outcomes align with intended objectives.

目錄
論文審定書.................................................................... i
摘要...........................................................................ii
Abstract......................................................................iii
目錄..........................................................................iv
圖次...........................................................................v
第一章緒論.................................................................1
第二章文獻探討............................................................6
第三章研究方法...........................................................10
3.1消費者行為.............................................................11
3.2創作者行為.............................................................15
3.3 AI公司行為............................................................16
3.4政府干預...............................................................18
第四章研究結果...........................................................20
4.1創作者誘因分析........................................................21
4.2 AI公司誘因分析........................................................23
4.3利潤交集變化與政策限制...............................................24
4.4社會福利分析...........................................................26
4.4.1區域定義與社會福利構成.........................................26
4.4.2制度選擇與社會福利比較.........................................29
第五章結論................................................................36
參考文獻.....................................................................38

參考文獻
Becker, G. S. (1968). Crime andpunishment: Aneconomicapproach. JournalofPolitical
Economy, 76(2):169–217.
Bick, A., Blandin, A., and Deming, D. J. (2024). The rapid adoption of generative AI.
Technical report, National Bureau of Economic Research.
Chellappa, R. K. and Shivendu, S. (2005). Managing piracy: Pricing and sampling strate
gies for digital experience goods in vertically segmented markets. Information Systems
Research, 16(4):400–417.
Chen, Y. N. and Png, I. (2003). Information goods pricing and copyright enforcement:
Welfare analysis. Information Systems Research, 14(1):107–123.
Eisend, M. (2019). Explaining digital piracy: A meta-analysis. Information Systems
Research, 30(2):636–664.
Guo, L. and Meng, X.(2015). Digital content provision and optimal copyright protection.
Management Science, 61(5):1183–1196.
Polinsky, A. M. and Shavell, S. (2000). The economic theory of public enforcement of
law. Journal of Economic Literature, 38(1):45–76.
Shavell, S. (1993). The optimal structure of law enforcement. The Journal of Law and
Economics, 36(1, Part 2):255–287.
Shy, O. and Thisse, J.-F. (1999). A strategic approach to software protection. Journal of
Economics & Management Strategy, 8(2):163–190.
Tunca, T. I. and Wu, Q. (2013). Fighting fire with fire: Commercial piracy and the role
of file sharing on copyright protection policy for digital goods. Information Systems
Research, 24(2):436–453.
Wang, J. T., Deng, Z., Chiba-Okabe, H., Barak, B., and Su, W. J. (2024). An economic
solution to copyright challenges of generative AI. arXiv preprint arXiv:2404.13964.