Üretken Yapay Zekânın Normatif Sınırları: Kapalı Kaynaklı ve Açık Kaynaklı Modellerde İçerik Kısıtlamalarının Karşılaştırmalı Bir İncelemesi

Öz

Üretken yapay zekâ sistemleri, üç yıldan kısa bir süre içinde acil hukuki, etik ve teknik kaygıların nesnesi hâline gelmiştir. Bu makale, kapalı kaynaklı ve açık kaynaklı üretken yapay zekâ modellerinin çıktıları üzerine konulan normatif sınırların karşılaştırmalı bir analizini sunmaktadır. Çalışma; içerik kısıtlamalarının iki yayın paradigması boyunca uygulanmasını sağlayan mimari, sözleşmesel ve yasal mekanizmaları karşılaştırmaktadır. Kapalı kaynaklı modellerde normatif yetki, API kapısı, RLHF ve tek taraflı kullanım politikaları yoluyla geliştiricilerde toplanmaktadır. Açık kaynaklı modeller ise ağırlıkları kamuya açıldıktan sonra uygulanabilirliği zayıflayan kullanım kısıtlı lisanslara dayanmaktadır. AB Yapay Zekâ Yasası, Çin Geçici Tedbirleri ve Türkiye Ulusal Yapay Zekâ Stratejisi 2021-2025 bu asimetrileri risk temelli araçlarla kapatmaya çalışmaktadır. Makale, en acil normatif meselenin her iki paradigmada meşruiyeti koruyabilecek çok merkezli yönetişim mimarilerinin inşası olduğunu savunmaktadır.

Anahtar Kelimeler

• AB Yapay Zekâ Yasası
• yapay zekâ yönetişimi
• kapalı kaynaklı modeller
• açık kaynaklı modeller
• içerik denetimi
• büyük dil modelleri
• üretken yapay zekâ

Etik Beyan

Bu makale, insan veya hayvan denekler içeren herhangi bir çalışma içermemektedir.

Normative Limits of Generative AI: A Comparative Study of Content Restrictions in Closed-Source and Open-Source Models

Öz

Generative artificial intelligence systems have become objects of urgent legal, ethical, and technical concern. This article offers a comparative analysis of the normative limits placed on the outputs of closed-source and open-source generative AI models. Drawing on literature spanning information technology law, platform governance, and comparative regulatory studies, the study contrasts the architectural, contractual, and statutory mechanisms by which content restrictions are operationalised across the two release paradigms. In closed-source models, normative authority sits with developers who enforce restrictions through API gating, Reinforcement Learning from Human Feedback (RLHF), and usage policies. Open-source models rely on use-restricted licences whose enforceability collapses once weights are publicly disseminated. Statutory regimes such as Regulation (EU) 2024/1689, China's Interim Measures for Generative AI Services, and Türkiye's National AI Strategy 2021-2025 attempt to bridge these asymmetries through risk-based instruments, yet remain calibrated to centralised compute. The article argues that the most pressing normative challenge is the construction of polycentric, multi-stakeholder governance architectures that preserve legitimacy across both paradigms.

Anahtar Kelimeler

• generative artificial intelligence
• large language models
• content moderation
• open-source models
• closed-source models
• AI governance
• EU AI Act

Etik Beyan

This article does not contain any studies with human or animal subjects.

Kaynakça

1. Akkurt, S. S. (2019). Yapay zekânın otonom davranışlarından kaynaklanan hukukî sorumluluk [Legal liability arising from the autonomous behaviour of artificial intelligence]. Uyuşmazlık Mahkemesi Dergisi, 7(13), 39–59. https://doi.org/10.18771/mdergi.581875
2. Bai, Y., Kadavath, S., Kundu, S., Askell, A., Kernion, J., Jones, A., Chen, A., Goldie, A., Mirhoseini, A., McKinnon, C., Chen, C., Olsson, C., Olah, C., Hernandez, D., Drain, D., Ganguli, D., Li, D., Tran-Johnson, E., Perez, E., . . . Kaplan, J. (2022). Constitutional AI: Harmlessness from AI feedback (arXiv:2212.08073). arXiv. https://doi.org/10.48550/arXiv.2212.08073
3. Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021, March). On the dangers of stochastic parrots: Can language models be too big?. In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency (pp. 610-623). https://doi.org/10.1145/3442188.3445922
4. Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., Brynjolfsson, E., Buch, S., Card, D., Castellon, R., Chatterji, N., Chen, A., Creel, K., Davis, J. Q., Demszky, D., . . . Liang, P. (2021). On the opportunities and risks of foundation models. arXiv. https://doi.org/10.48550/arXiv.2108.07258
5. Chen, C., Qu, W., Su, S., Feng, Y., & Li, T. (2025). A comprehensive review of LLM-based content moderation: Advancements, challenges, and future directions. Knowledge-Based Systems, 330, 114689. https://doi.org/10.1016/j.knosys.2025.114689
6. China Law Translate. (2023, July 13). Interim Measures for the Management of Generative Artificial Intelligence Services (Translation). https://www.chinalawtranslate.com/en/generativeai-interim/
7. Cui, J., & Araujo, D. A. (2024). Rethinking use-restricted open-source licenses for regulating abuse of generative models. Big Data & Society, 11(1), 20539517241229699. https://doi.org/10.1177/20539517241229699
8. Demirtaş, A. (2024). ChatGPT’nin gölgesinde kişisel verilerin korunması [Protection of personal data in the shadow of ChatGPT]. Kişisel Verileri Koruma Dergisi, 6(1), 14–27. https://izlik.org/JA54PN97RK

9. Digital Transformation Office of the Presidency of the Republic of Türkiye, & Ministry of Industry and Technology. (2021). Ulusal Yapay Zekâ Stratejisi 2021–2025 [National Artificial Intelligence Strategy 2021–2025]. Presidential Circular No. 2021/18, Official Gazette of the Republic of Türkiye (No. 31574). cbddo.gov.tr
10. Dönmez, Z. (2025). Yapay zekânın hukuki gelişimi ve AB yapay zekâ yasası [Legal development of artificial intelligence and the EU Artificial Intelligence Act]. Adalet Dergisi, 74, 947–975. https://doi.org/10.57083/adaletdergisi.1677361
11. European Commission. (2025). The general-purpose AI code of practice. https://digital-strategy.ec.europa.eu/en/policies/contents-code-gpai
12. Ferrari, F., Van Dijck, J., & Van den Bosch, A. (2025). Observe, inspect, modify: Three conditions for generative AI governance. New Media & Society, 27(5), 2788-2806. https://doi.org/10.1177/14614448231214811
13. Ganguli, D., Lovitt, L., Kernion, J., Askell, A., Bai, Y., Kadavath, S., Mann, B., Perez, E., Schiefer, N., Ndousse, K., Jones, A., Bowman, S., Chen, A., Conerly, T., DasSarma, N., Drain, D., Elhage, N., El-Showk, S., Fort, S., . . . Clark, J. (2022). Red teaming language models to reduce harms: Methods, scaling behaviors, and lessons learned (arXiv:2209.07858v2). arXiv. https://doi.org/10.48550/arXiv.2209.07858
14. Gao, L., Ahmed, A., Chen, O., Reyl, M., Cheema, Z., Feamster, N., ... & Chetty, M. (2026, April). Governance of AI-generated content: A case study on social media platforms. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (pp. 1-22). https://doi.org/10.1145/3772318.3790415
15. Gill, A. S., & Germann, S. (2022). Conceptual and normative approaches to AI governance for a global digital ecosystem supportive of the UN Sustainable Development Goals (SDGs). AI and Ethics, 2(2), 293–301. https://doi.org/10.1007/s43681-021-00058-z
16. Gorwa, R., Binns, R., & Katzenbach, C. (2020). Algorithmic content moderation: Technical and political challenges in the automation of platform governance. Big Data & Society, 7(1), 1–15. https://doi.org/10.1177/2053951719897945
17. Hacker, P., Engel, A., & Mauer, M. (2023). Regulating ChatGPT and other large generative AI models. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency (pp. 1112–1123). Association for Computing Machinery. https://doi.org/10.1145/3593013.3594067
18. Hao, S., Kumar, P., Laszlo, S., Poddar, S., Radharapu, B., & Shelby, R. (2023). Safety and fairness for content moderation in generative models. arXiv preprint arXiv:2306.06135. https://doi.org/10.48550/arXiv.2306.06135 Hirschman, A. O. (1970). Exit, voice, and loyalty: Responses to decline in firms, organizations, and states. Harvard University Press.
19. Huang, T. (2025). Content moderation by LLM: From accuracy to legitimacy. Artificial Intelligence Review, 58(8), 320. https://doi.org/10.1007/s10462-025-11328-1
20. Karadeniz, S. (2025). Avrupa Birliği Yapay Zekâ Kanunu’nun risk grupları ve ilgililerin yükümlülükleri bağlamında incelenmesi [An analysis of the European Union Artificial Intelligence Act in the context of risk groups and the obligations of related parties]. Ankara Hacı Bayram Veli Üniversitesi Hukuk Fakültesi Dergisi, 29(1), 307–366. https://doi.org/10.34246/ahbvuhfd.1541455
21. Kişisel Verileri Koruma Kurumu. (2025). Yapay zekâ alanında kişisel verilerin korunmasına dair tavsiyeler [Recommendations on the protection of personal data in the field of artificial intelligence] (KVKK Publications No. 76). https://www.kvkk.gov.tr/Icerik/7048/Yapay-Zeka-Alaninda-Kisisel-Verilerin-Korunmasina-Dair-Tavsiyeler
22. Kişisel Verileri Koruma Kurumu. (2025, November 24). Üretken yapay zekâ ve kişisel verilerin korunması rehberi (15 soruda) [Guide on generative artificial intelligence and the protection of personal data (in 15 questions)]. https://www.kvkk.gov.tr/Icerik/8547/uretken-yapay-zeka-ve-kisisel-verilerin-korunmasi-rehberi-15-soruda
23. Kişisel Verilerin Korunması Kanunu [Law on the Protection of Personal Data], Law No. 6698. (2016, April 7). Resmî Gazete (Issue No. 29677).
24. Kumayama, K. D., & Chiruvolu, P. (2024, August 15). Legal primer on open genAI models. Reuters Legal News. https://www.reuters.com/legal/legalindustry/legal-primer-open-genai-models-2024-08-15/
25. Law on the Regulation of Publications on the Internet and Combating Crimes Committed by Means of Such Publications (Law No. 5651). (2007, May 23). Resmî Gazete (No. 26530).
26. Machado, J. (2025). Toward a public and secure generative AI: A comparative analysis of open and closed LLMs. arXiv. https://doi.org/10.48550/arXiv.2505.10603
27. Mueller, A. R., & Yoo, C. S. (2026). Taking standards seriously: The case for a private standards-based approach to AI governance. Berkeley Technology Law Journal, 40(4), 1273–1314. https://doi.org/10.15779/Z38WS8HP0D
28. Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., Schulman, J., Hilton, J., Kelton, F., Miller, L., Simens, M., Askell, A., Welinder, P., Christiano, P., Leike, J., & Lowe, R. (2022). Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35, 27730–27744. https://proceedings.neurips.cc/paper/2022/hash/b1efde53be364a73914f58805a001731-Abstract-Conference.html
29. Özparlak, B. O., & Çetin, M. (2023). ChatGPT ve üretici yapay zekâ modellerinde mahremiyet ve güvenliğin hukuki boyutu [The legal dimension of privacy and security in ChatGPT and generative artificial intelligence models]. Marmara Üniversitesi Hukuk Fakültesi Hukuk Araştırmaları Dergisi, 29(2), 1003–1040. https://doi.org/10.33433/maruhad.1347497
30. Qi, X., Zeng, Y., Xie, T., Chen, P. Y., Jia, R., Mittal, P., & Henderson, P. (2024). Fine-tuning aligned language models compromises safety, even when users do not intend to! International Conference on Learning Representations. https://openreview.net/forum?id=hTEGyKf0dZ
31. Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act). (2024). Official Journal of the European Union, L 2024/1689. http://data.europa.eu/eli/reg/2024/1689/oj
32. Reuel, A., & Undheim, T. A. (2024). Generative AI needs adaptive governance. arXiv preprint arXiv:2406.04554. https://doi.org/10.48550/arXiv.2406.04554
33. Scharpf, F. W. (1999). Governing in Europe: Effective and democratic? Oxford University Press. Sokhansanj, B. A. (2025). Local ai governance: Addressing model safety and policy challenges posed by decentralized ai. AI, 6(7), 159. https://doi.org/10.3390/ai6070159
34. Solaiman, I. (2023). The gradient of generative AI release: Methods and considerations. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency (pp. 111–122). Association for Computing Machinery. https://doi.org/10.1145/3593013.3593981
35. Solaiman, I., Brundage, M., Clark, J., Askell, A., Herbert-Voss, A., Wu, J., ... & Wang, J. (2019). Release strategies and the social impacts of language models. arXiv preprint arXiv:1908.09203. https://doi.org/10.48550/arXiv.1908.09203
36. Suzor, N. P. (2019). Lawless: The secret rules that govern our digital lives. Cambridge University Press.
37. Türk Borçlar Kanunu [Turkish Code of Obligations] (Law No. 6098). (2011, February 4). Resmî Gazete (No. 27836).
38. Weidinger, L., Mellor, J., Rauh, M., Griffin, C., Uesato, J., Huang, P.-S., Cheng, M., Glaese, M., Balle, B., Kasirzadeh, A., Kenton, Z., Brown, S., Hawkins, W., Stepleton, T., Biles, C., Birhane, A., Haas, J., Rimell, L., Hendricks, L. A., . . . Gabriel, I. (2021). Ethical and social risks of harm from language models. arXiv. https://doi.org/10.48550/arXiv.2112.04359
39. Weidinger, L., Uesato, J., Rauh, M., Griffin, C., Huang, P.-S., Mellor, J., Glaese, A., Cheng, Myra., Balle, B., Kasirzadeh, A., Biles, C., Brown, S., Kenton, Z., Hawkins, W., Stepleton, T., Birhane, A., Hendricks, L. A., Rimell, L., Isaac, W., . . . Gabriel, I. (2022). Taxonomy of risks posed by language models. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (pp. 214-229). Association for Computing Machinery. https://doi.org/10.1145/3531146.3533088
40. Yamak, Y. (2025). Yapay zekâ ile ilgili güncel düzenleme çalışmaları: AB, ABD ve Türkiye açısından incelenmesi [Current regulatory studies on artificial intelligence: A review from the perspective of the EU, USA, and Türkiye]. Bilgi Teknolojileri ve İletişim Dergisi, 3(1), 35–68. https://izlik.org/JA36XD67BP
41. Yang, W., He, R., & Zhou, M. (2026). Beyond banning AI: A first look at GenAI governance in open-source software communities. arXiv. https://doi.org/10.48550/arXiv.2603.26487
42. Yünlü, S. (2024). Üretici yapay zekâ kaynaklı norm ve kişi bazlı hukuki sorumluluk [Norm and person-based legal liability arising from generative artificial intelligence]. Adalet Dergisi, 72, 501–542. https://doi.org/10.57083/adaletdergisi.1484067
43. Zhang, H., Edelman, B. L., Francati, D., Venturi, D., Ateniese, G., & Barak, B. (2023). Watermarks in the sand: Impossibility of strong watermarking for generative models. arXiv. https://doi.org/10.48550/arXiv.2311.04378