THE PROBLEM OF BIAS IN ARTIFICIAL INTELLIGENCE DECISIONS AND WAYS TO ELIMINATE IT
DOI:
https://doi.org/10.47390/SPR1342V6I2Y2026N30Keywords:
artificial intelligence, fairness, bias, ethical issues, algorithmic justice.Abstract
This study explores how unfair outcomes emerge within artificial intelligence decision-making and investigates the underlying factors that contribute to such bias. Since AI models rely heavily on human-generated data, any distortions, stereotypes, or inaccuracies present in those datasets can be reproduced and amplified by the system. The article examines the roots of biased decisions, evaluates their consequences for individuals and society, and identifies practical approaches for reducing these risks. The proposed solutions include technical measures – such as refining datasets and applying fairness-oriented algorithms – as well as ethical strategies that involve strengthening transparency mechanisms and establishing oversight bodies. The results indicate that ensuring fairness in AI-based decisions requires not only technological improvements but also careful consideration of human involvement throughout the entire development process.
References
1. Crawford K. (2021). Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. Yale University Press.
2. Russell S., Norvig P. (2021). Artificial Intelligence: A Modern Approach. Pearson Education.
3. The Conversation (2024). “How to prevent AI bias: Lessons from recent AI failures.” Available at: https://theconversation.com
4. European Commission (2024). Artificial Intelligence Act (EU Regulation 2024/1688). Brussels: Official Journal of the European Union.
5. Binns R. (2018). “Fairness in machine learning: Lessons from political philosophy.” Proceedings of the FAT Conference. New York: ACM.
6. Giddens A. (1984). The Constitution of Society: Outline of the Theory of Structuration. University of California Press.
7. ProPublica (2016). “Machine Bias: Risk Assessments in Criminal Sentencing.” Available at: https://www.propublica.org
8. UNESCO (2021). Recommendation on the Ethics of Artificial Intelligence. Paris: UNESCO Press.
9. Winfield A. F. T., Jirotka M. (2018). “Ethical governance is essential to building trust in robotics and AI.” Philosophical Transactions of the Royal Society A.
10. Reuters (2018). “Amazon scraps AI recruiting tool that showed bias against women.” Reuters Technology News.
11. OECD (2022). Recommendation on AI Governance. Paris: OECD Publishing.
12. O‘zbekiston Respublikasi Prezidentining 2024-yil 14-oktabrdagi «Sunʼiy intellekt texnologiyalarini 2030-yilga qadar rivojlantirish strategiyasini tasdiqlash toʻgʻrisida»gi PQ-358-son qarori.
