Kamu Yönetiminin Şeffaflığından Algoritmaların Şeffaflığına Geçiş: Şeffaflığı Algoritmaların Opaklığı ve Kara Kutu Problemi Çerçevesinde Yeniden Değerlendirmek

Öz

Bu çalışma, kamu yönetiminde şeffaflık kavramının, yapay zeka (YZ) ve algoritmik sistemlerin yaygın kullanımıyla nasıl yeniden şekillendiğini analiz etmeyi amaçlamaktadır. Algoritmaların iç işleyişinin anlaşılmasını zorlaştıran ‘opaklık’ ve ‘kara kutu problemi’, geleneksel şeffaflık anlayışını yetersiz hale getirmekte ve karar alma süreçlerinin daha ayrıntılı bir şekilde incelenmesini gerektirmektedir. Çalışmada, YZ tabanlı sistemlerin kamu yönetiminde kullanımının doğurduğu etik, sosyal ve hukuki meseleler ele alınarak, ‘açıklanabilir YZ’ (Explainable AI - XAI) ve ‘açık kaynak politikaları’ gibi çözüm önerileri tartışılmaktadır. Elde edilen bulgular, algoritmik şeffaflığın teknik kısıtlar, veri güvenliği riskleri ve vatandaşların karar süreçlerini anlama kapasitesi gibi çeşitli sınırlılıklarla karşı karşıya olduğunu ortaya koymaktadır. Sonuç olarak, algoritmik sistemlerin şeffaflığı, demokratik değerler ve hesap verebilirlik açısından kritik öneme sahip olup, bu alanda çok boyutlu ve bütüncül yaklaşımlar geliştirilmesi gerekmektedir.

Anahtar Kelimeler

• Kamu yönetimi
• yapay zeka
• algoritma
• şeffaflık
• hesap verebilirlik

Transition from Transparency in Public Administration to Transparency in Algorithms: Re-evaluating Transparency within the Framework of Algorithmic Opacity and the Black Box Problem

Abstract

This study aims to analyze how the concept of transparency in public administration is being reshaped by the widespread use of artificial intelligence (AI) and algorithmic systems. The 'opacity' and 'black box problem,' which make it difficult to understand the inner workings of algorithms, render traditional notions of transparency inadequate and necessitate a more detailed examination of decision-making processes. The study addresses the ethical, social, and legal issues arising from the use of AI-based systems in public administration and discusses potential solutions such as 'Explainable AI' (XAI) and 'open-source policies.' The findings reveal that algorithmic transparency faces various limitations, including technical constraints, data security risks, and the public's capacity to understand decision-making processes. In conclusion, the transparency of algorithmic systems is of critical importance for democratic values and accountability, requiring the development of multidimensional and holistic approaches in this field.

Keywords

• Public administration
• artificial intelligence
• algorithm
• transparency
• accountability

Kaynakça

1. Adadi, A., ve Berrada, M. (2018), Peeking Inside The Black-Box: A Survey On Explainable Artificial Intelligence (XAI), IEEE Access, 6, s.52138-52160.
2. Ali, S., Abuhmed, T., El-Sappagh, S., Muhammad, K., Alonso-Moral, J. M., Confalonieri, R., ... ve Herrera, F. (2023), Explainable Artificial Intelligence (XAI): What We Know And What Is Left To Attain Trustworthy Artificial Intelligence, Information Fusion, 99, s.1-52.
3. Aneesh, A. (2002), Technologically Coded Authority: The Post-Industrial Decline in Bureaucratic Hierarchies, 7th International Summer Academy on Technology Studies, Deutschlandsberg, Austria, s.1-25.
4. Aneesh, A. (2009), Global Labor: Algocratic Modes of Organization, Sociological theory, 27(4), s.347-370.
5. Angwin, J., Larson, J., Mattu, S., ve Kirchner, L. (2016), Machine Bias. There’s Software Used Across The Country to Predict Future Criminals. And It’s Biased Against Blacks. https://www.propublica.org/article/machine-bias-risk- assessments-in-criminal-sentencing (22.01.2025)
6. Arrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., ... ve Herrera, F. (2020), Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities And Challenges Toward Responsible AI, Information Fusion, 58, s.82-115.
7. Barocas, S., ve Selbst, A. D. (2016), Big Data's Disparate Impact, California Law Review, 104(3), s.671-732.
8. Bouwmeester, M. (2023), System Failure in The Digital Welfare State: Exploring Parliamentary And Judicial Control in The Dutch Childcare Benefits Scandal, Recht der Werkelijkheid, 44(2), s.13-37.

9. Bovens, M., Schillemans, T., ve Hart, P. T. (2008), Does Public Accountability Work? An Assessment Tool, Public Administration, 86(1), s.225-242.
10. Brauneis, R., ve Goodman, E. P. (2017), Algorithmic Transparency For The Smart City, Yale Journal of Law & Technology, 20, s.104-176.
11. Brundage, M., Avin, S., Clark, J., Toner, H., Eckersley, P., Garfinkel, B., ... ve Amodei, D. (2018), The Malicious Use of Artificial Intelligence: Forecasting, Prevention, And Mitigation. https://arxiv.org/pdf/1802.07228 (15.01.2025)
12. Burrell, J. (2016), How The Machine ‘Thinks’: Understanding Opacity in Machine Learning Algorithms, Big data & society, 3(1), s.1-12.
13. Chowdhury, S. (2024), Technology Is Never Neutral: Robodebt And A Human Rights Analysis of Automated Decision-Making on Welfare Recipients, Australian Journal of Human Rights, 30(1), s.20-40.
14. Christie, J. (2022), The Post Office IT Scandal-Why IT Audit Is Essential for Effective Corporate Governance, Digital Evidence & Elec. Signature Law Review, 19, s.42-86.
15. Citron, D. K., ve Pasquale, F. (2014), The Scored Society: Due Process for Automated Predictions, Washington Law Reviwe, 89, s.1-34.
16. Coglianese, C., ve Lehr, D. (2019), Transparency And Algorithmic Governance, Administrative law review, 71(1), s.1-56.
17. Danaher, J. (2016), The Threat of Algocracy: Reality, Resistance And Accommodation, Philosophy & Technology, 29(3), s.245-268.
18. Diakopoulos, N. (2015), Algorithmic Accountability: Journalistic Investigation of Computational Power Structures, Digital Journalism, 3(3), s.398-415.
19. Edwards, L., ve Veale, M. (2017), Slave to The Algorithm? Why A 'Right to An Explanation' Is Probably Not The Remedy You Are Looking for, Duke Law & Technology Review, 16, s.18-84.
20. Erkkilä, T. (2020), Transparency In Public Administration. Oxford Research Encyclopedia of Politics. Oxford: Oxford University Press. https://oxfordre.com/politics/display/10.1093/acrefore/9780190228637.001.0001/acrefore- 9780190228637-e-1404 (17.01.2025)
21. Etscheid, J. (2019), Artificial Intelligence in Public Administration: A Possible Framework for Partial And Full Automation, In Electronic Government: 18th IFIP WG 8.5 International Conference, EGOV 2019, San Benedetto Del Tronto.
22. Etzioni, A., ve Etzioni, O. (2016), Keeping AI Legal, Vanderbilt Journal of Entertainment & Technology Law, 19(1), s.133-146.
23. Fink, K. (2018), Opening The Government’s Black Boxes: Freedom of İnformation And Algorithmic Accountability, Information, Communication & Society, 21(10), s.1453-1471.
24. Gandomi, A., ve Haider, M. (2015), Beyond The Hype: Big Data Concepts, Methods, And Analytics, International Journal of Information Management, 35(2), s.137-144.
25. Gillespie, T. (2014), The relevance of algorithms. http://governingalgorithms.org/wp- content/uploads/2013/05/1-paper-gillespie.pdf (11.01.2025)
26. Hadwick, D., & Lan, S. (2021), Lessons to Be Learned from The Dutch Childcare Allowance Scandal: Comparative Review of Algorithmic Governance By Tax Administrations in The Netherlands, France And Germany, World Tax Journal, 2021(4), s.609-645.
27. Heaton, D., Nichele, E., Clos, J., ve Fischer, J. E. (2023), “The Algorithm Will Screw You”: Blame, Social Actors And The 2020 A Level Results Algorithm on Twitter, Plos One, 18(7), s.1-29.
28. Hood, C. (2007). What Happens When Transparency Meets Blame-Avoidance?. Public Management Review, 9(2), 191-210.
29. Kamilaris, A., Kartakoullis, A., ve Prenafeta-Boldú, F. X. (2017), A Review on The Practice of Big Data Analysis in Agriculture, Computers and Electronics in Agriculture, 143, s.23-37.
30. Kolkman, D. (2020). F** K The Algorithm?: What The World Can Learn From The UK’s A-Level Grading Fiasco. Impact of Social Sciences Blog, 1-4. https://blogs.lse.ac.uk/impactofsocialsciences/2020/08/26/fk-the- algorithm-what-the-world-can-learn-from-the-uks-a-level-grading-fiasco/ (17.01.2025).
31. LeCun, Y., Bengio, Y., ve Hinton, G. (2015), Deep Learning, Nature, 521(7553), s.436-444.
32. Lepri, B., Oliver, N., Letouzé, E., Pentland, A., ve Vinck, P. (2018), Fair, Transparent, And Accountable Algorithmic Decision-Making Processes: The Premise, The Proposed Solutions, And The Open Challenges, Philosophy & Technology, 31(4), s.611-627.
33. Levy, K., Chasalow, K. E., ve Riley, S. (2021), Algorithms And Decision-Making in The Public Sector, Annual Review of Law and Social Science, 17(1), s.309-334.
34. Lipton, Z. C. (2018), The Mythos of Model Interpretability: In Machine Learning, The Concept of Interpretability Is Both Important And Slippery, Queue, 16(3), s.31-57.
35. Margetts, H., ve Dorobantu, C. (2019), Rethink Government with AI, Nature, 568(7751), s.163-165.
36. Marshall, P. (2022), Scandal At The Post Office: The Intersection of Law, Ethics And Politics, Digital Evidence & Elec. Signature Law Review, 19, s.12-28.
37. Mayer-Schönberger, V., ve Cukier, K. (2013), Big Data: A Revolution That Will Transform How We Live, Work, And Think, New York: Houghton Mifflin Harcourt.
38. Meijer, A. (2013), Understanding The Complex Dynamics of Transparency, Public Administration Review, 73(3), s.429-439.
39. O'neil, C. (2017), Weapons of Math Destruction: How Big Data Increases Inequality And Threatens Democracy, New York: Crown Publishers.
40. Pasquale, F. (2015), The Black Box Society: The Secret Algorithms That Control Money And Information, Cambridge, Massachusetts: Harvard University Press.
41. Rudin, C. (2019), Stop Explaining Black Box Machine Learning Models for High Stakes Decisions And Use Interpretable Models Instead, Nature Machine Intelligence, 1(5), s.206-215.
42. Rudin, C., Wang, C., ve Coker, B. (2020), The Age of Secrecy And Unfairness in Recidivism Prediction. Harvard Data Science Review, 2(1), s.1-55.
43. Selbst, A. D., ve Barocas, S. (2018), The İntuitive Appeal of Explainable Machines, Fordham Law Review, 87, s.1085- 1140.
44. Sobrino-García, I. (2021), Artificial Intelligence Risks And Challenges in The Spanish Public Administration: An Exploratory Analysis Through Expert Judgements, Administrative Sciences, 11(3), s.1-23.
45. Sun, T. Q., ve Medaglia, R. (2019), Mapping The Challenges of Artificial Intelligence in The Public Sector: Evidence from Public Healthcare, Government Information Quarterly, 36(2), s.368-383.
46. Thierer, A., O’Sullivan, A., ve Russell, R. (2017), Artificial Intelligence And Public Policy, Arlington, VA: Mercatus Center George Mason University.
47. Von Eschenbach, W. J. (2021), Transparency And The Black Box Problem: Why We Do Not Trust AI, Philosophy & Technology, 34(4), s.1607-1622.
48. Welch, E. W., Hinnant, C. C., ve Moon, M. J. (2005), Linking Citizen Satisfaction with E-Government And Trust in Government, Journal of Public Administration Research And Theory, 15(3), s.371-391.
49. West, D. M., ve Allen, J. R. (2018), How Artificial Intelligence Is Transforming The World, Brookings Institution. https://www.brookings.edu/articles/how-artificial-intelligence-is-transforming-the-world/ (15.01.2015)
50. Whiteford, P. (2021), Debt By Design: The Anatomy of A Social Policy Fiasco–Or Was It Something Worse?, Australian Journal of Public Administration, 80(2), s.340-360.
51. Wirtz, B. W., Weyerer, J. C., ve Geyer, C. (2019), Artificial Intelligence And The Public Sector—Applications And Challenges, International Journal of Public Administration, 42(7), s.596-615.
52. Wirtz, B. W., Weyerer, J. C., ve Sturm, B. J. (2020), The Dark Sides of Artificial Intelligence: An Integrated AI Governance Framework for Public Administration, International Journal of Public Administration, 43(9), s.818- 829.
53. Zerilli, J., Knott, A., Maclaurin, J., ve Gavaghan, C. (2019), Transparency in Algorithmic And Human Decision- Making: Is There A Double Standard?, Philosophy & Technology, 32, s.661-683.