Ku etycznej sztucznej inteligencji. Jak uwzględniać uprzedzenia, prywatność i odpowiedzialność w systemach opartych na danych
DOI:
https://doi.org/10.34767/SIIP.2025.01.04Słowa kluczowe:
etyka AI, systemy oparte na danych, uprzedzenia algorytmiczne, prywatność, odpowiedzialność, etyczne zarządzanieAbstrakt
Niniejszy artykuł analizuje implikacje etyczne związane ze sztuczną inteligencją i systemami przetwarzającymi dane, koncentrując się na tym, w jaki sposób kwestie związane z uprzedzeniami, prywatnością oraz bezpieczeństwem wpływają na wdrażanie technologii AI w różnych obszarach. Głównym problemem jest tu tendencja algorytmów AI do wzmacniania uprzedzeń społecznych i technicznych, co może prowadzić do ograniczenia sprawiedliwości oraz szkodzić grupom marginalizowanym. Warto jednak zauważyć, że decyzje podejmowane przez algorytmy i systemy AI mogą również przyczynić się do redukcji uprzedzeń, jeśli zapewniona zostanie transparentność i oparcie na faktach, pod warunkiem, że systemy te powstają w sposób etyczny. Aby zbadać te zagadnienia, zastosowano podejście koncepcyjne i jakościowe, łącząc wnioski z istniejących ram regulacyjnych, studiów przypadków oraz literatury naukowej dotyczącej etyki AI i analityki danych. Podstawowa hipoteza głosi, że integracja zasad etycznych, rzetelnego nadzoru oraz współpracy między różnymi dyscyplinami w procesie projektowania i rozwoju systemów AI pozwala na ograniczanie szkodliwych uprzedzeń, ochronę prywatności i wzmacnianie zaufania społecznego. Hipotezę tę weryfikuje się poprzez analizę ról różnych interesariuszy – twórców systemów AI, decydentów politycznych i użytkowników końcowych – w powstawaniu bądź redukowaniu stronniczości w procesach algorytmicznych. W konkluzji artykuł wskazuje na konieczność ustanowienia kompleksowych standardów regulacyjnych, większej przejrzystości działania modeli oraz ich regularnych aktualizacji, przy jednoczesnym zaangażowaniu zarówno ekspertów, jak i ogółu społeczeństwa. Tylko w ten sposób technologie AI będą mogły funkcjonować w sposób zgodny z wartościami demokratycznymi i humanistycznymi.
Bibliografia
Almutairi, Z., & Elgibreen, H. (2022). A Review of Modern Audio Deepfake Detection Methods: Challenges and Future Directions. Algorithms, 15(5), 155. https://doi.org/10.3390/a15050155
Alpaydin, E. (2016). Machine Learning: The New AI. Cambridge Mass.: MIT Press.
Angwin, J. (2016). Sample: COMPAS Risk Assessment, COMPAS “CORE”. DocumentCloud. https://www.documentcloud.org/documents/2702103-Sample-Risk-Assessment- COMPAS-CORE (accessed: 16.03.2023).
Angwin, J., Larson, J., Mattu, S., & Kirchner, L. (2016, May 23). Machine bias. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing (accessed: 16.03.2023).
Baldridge, J. (2015, August 2). Machine learning and human bias: An uneasy pair. TechCrunch. https://techcrunch.com/2015/08/02/machine-learning-and-human-bias-an-uneasy-pair (accessed: 11.03.2025).
Boden, M.A. (2016). AI: Its Nature and Future. Oxford: Oxford University Press.
Bostrom, N. (2016). Superintelligence. Oxford: Oxford University Press.
Buolamwini, J., & Gebru, T. (2018). Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification. Proceedings of Machine Learning Research, 81, 1–15.
Cataleta, M.S. (2020). Humane Artificial Intelligence: The Fragility of Human Rights Facing AI. East-West Center. http://www.jstor.org/stable/resrep25514
Chesney, R., & Citron, D. (2019). Deepfakes and the New Disinformation War: The Coming Age of Post-Truth Geopolitics. Foreign Affairs, 98(1), 147–155.
Chun, W.H.K. (2021). Discriminating Data: Correlation, Neighborhoods, and the New Politics of Recognition. Cambridge Mass.: The MIT Press.
Coeckelbergh, M. (2020). AI Ethics. Cambridge Mass: The MIT Press.
Collins, A., & Ebrahimi, T. (2021). Risk governance and the rise of deepfakes. International Risk Governance Center, 1–4. https://doi.org/10.5075/epflirgc-285637
Dressel, J., & Farid, H. (2018). The accuracy, fairness, and limits of predicting recidivism. Science Advances, 4(1), 1–5. https://doi.org/10.1126/sciadv.aao5580
D’Ignazio, C., & Klein, L.F. (2020). Data Feminism. Cambridge Mass.: The MIT Press.
European Commission AI HLEG (High-Level Expert Group on Artificial Intelligence) (2019). Ethics Guidelines for Trustworthy AI. Brussels, European Commission. https://ec.europa.eu/futurium/en/ai-alliance-consultation/guidelines (accessed: 12.04.2023).
Floridi, L. (2023). The Ethics of Artificial Intelligence: Principles, Challenges, and Opportunities. Oxford: Oxford University Press.
Fry, H. (2018). Hello World: How to Be Human in the Age of the Machine (1st ed.). London: Transworld Publishers.
Fuchs, C. (2013). Digital prosumption labour on social media in the context of the capitalist regime of time. Time & Society, 23(1), 97–123. https://doi.org/10.1177/0961463x13502117
Houser, K.A. (2019). Can AI Solve the Diversity Problem in the Tech Industry? Mitigating Noise and Bias in Employment Decision-Making. The Stanford Technology Law Review, 290, 290–354.
Howard, J. (2019). Artificial intelligence: Implications for the future of work. American Journal of Industrial Medicine, 62(11), 917–926. https://doi.org/10.1002/ajim.23037
IBM Data and AI Team (2023, October 12). Types of Artificial Intelligence. https://www.ibm.com/think/topics/artificial-intelligence-types (accessed: 8.06.2024).
IEEE (N/A.). General Principles. The IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems. https://standards.ieee.org/wp-content/uploads/import/documents/other/ead1e_general_principles.pdf (accessed: 16.03.2024).
Iman, M., Arabnia, H.R., & Maribe Branchinst, R. (2022). Pathways to Artificial General Intelligence: A Brief Overview of Developments and Ethical Issues via Artificial Intelligence, Machine Learning, Deep Learning, and Data Science. In: H.R. Arabnia et al. (eds.), Advances in Artificial Intelligence and Applied Cognitive Computing (pp. 73–87). Berlin: Springer Nature.
Jagtiani, J., & Lemieux, C. (2019, January). The Roles of Alternative Data and Machine Learning in Fintech Lending: Evidence from the LendingClub Consumer Platform. Working Paper (Federal Reserve Bank of Philadelphia). https://doi.org/10.21799/frbp.wp.2018.15
Jemielniak, D., & Przegalińska, A.K. (2020). Collaborative Society. Cambridge Mass.: MIT Press.
Kahneman, D., Sibony, O., & Sunstein, C.R. (2021). Noise: A Flaw in Human Judgment. New York, NY: Littlle, Brown Spark.
Kelleher, J.D., & Tierney, B. (2018). Privacy and Ethics. In: eidem, Data Science (pp. 181–218). Cambridge Mass.: MIT Press.
Kilbertus, N., Gascon, A., Kusner, M., Veale, M., Gummadi, K., & Weller, A. (2018). Blind Justice: Fairness with Encrypted Sensitive Attributes. Proceedings of the 35th International Conference on Machine Learning, Proceedings of Machine Learning Research, 80, 2630–2639. https://proceedings.mlr.press/v80/kilbertus18a.html (accessed: 20.05.2023).
Kronmal, R.A. (1993). Spurious Correlation and the Fallacy of the Ratio Standard Revisited. Journal of the Royal Statistical Society. Series A (Statistics in Society), 156(3), 379–392. https://doi.org/10.2307/2983064
Lipińska, I. (2022). Etyka sztucznej inteligencji w dokumentach Unii Europejskiej w latach 2017–2020. Edukacja Filozoficzna, 73, 11–38. https://doi.org/10.14394/edufil.2022.0001
Lopez, P. (2021). Bias Does Not Equal Bias: A Socio-Technical Typology of Bias in Data-Based Algorithmic Systems. Internet Policy Review, 10(4). https://doi.org/10.14763/2021.4.1598
Madon, S., Willard, J., Guyll, M., & Scherr, K.C. (2011). Self‐Fulfilling Prophecies: Mechanisms, Power, and Links to Social Problems. Social and Personality Psychology Compass, 5(8), 578–590. https://doi.org/10.1111/j.1751-9004.2011.00375.x
McIntyre, L.C. (2018). Post-truth. Cambridge Mass.: MIT Press.
Mamak-Zdanecka, M., Stojkow, M., & Żuchowska-Skiba, D. (2019). Społeczny wymiar algorytmizacji. Humanizacja Pracy, 3(297), 9–20.
Mazurek, G. (2023). Sztuczna inteligencja, prawo i etyka. Krytyka Prawa, 15(1), 7–10. https://doi.org/10.7206/kp.2080-1084.567
Mejias, U.A., & Couldry, N. (2019). Datafication. Internet Policy Review, 8(4). https://doi.org/10.14763/2019.4.1428
Microsoft (2018). The Future Computed: Artificial Intelligence and Its Role in Society. Redmond: Microsoft Corporation.
O’Neil, C. (2016). Weapons of Math Destruction. New York: Crown Publishing Group.
Rai, N. (2021). Why ethical audit matters in artificial intelligence? AI and Ethics, 2(1), 209–218. https://doi.org/10.1007/s43681-021-00100-0
Sadok, H., Sakka, F., & El Hadi El Maknouzi, M. (2022). Artificial Intelligence and Bank Credit Analysis: A Review. Cogent Economics & Finance, 10(1), 2023262. https://doi.org/10.1080/23322039.2021.2023262
Searle, J.R. (1983). Can Computers Think? In: idem, Minds, Brains, and Science (pp. 28–41). Cambridge Mass.: Harvard University Press.
Smith, B., & Browne, C.A. (2021). Tools and Weapons: The Promise and the Peril of the Digital Age. London: Penguin Books.
Tencent Keen Security Lab (2019). Experimental security research of Tesla Autopilot [Technical report]. https://keenlab.tencent.com/en/whitepapers/Experimental_Security_Research_of_Tesla_Autopilot.pdf (accessed: 11.03.2025).
Vigen, T. (2015). Spurious Correlations. New York: Hachette Books.
Whittaker, L., Letheren, K., & Mulcahy, R. (2021). The Rise of Deepfakes: A Conceptual Framework and Research Agenda for Marketing. Australasian Marketing Journal, 29(3), 204–214. https://doi.org/10.1177/1839334921999479
