ETHICAL ISSUES AND THE INTERPRETABILITY CRISIS IN ARTIFICIAL INTELLIGENCE: A HUMAN-CENTERED APPROACH TO TRUST AND ACCOUNTABILITY

Salman A.

6th INTERNATIONAL ETHICS CONGRESS, Ankara, Türkiye, 8 - 09 Aralık 2025, ss.150-160, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Ankara
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.150-160
  • Maltepe Üniversitesi Adresli: Evet

Özet

Artificial intelligence (AI) has, over the last decade, moved beyond being merely a technical innovation and has become the centerpiece of a multilayered transformation affecting nearly all domains of human life. Across a wide spectrum from healthcare and education to financial markets, public administration, social media, and intimate relationships AI systems are automating decision-making processes and increasingly substituting human judgment. This rapid transformation extends beyond efficiency and innovation; it has also created an unprecedented sphere of ethical, legal, and social consequences in human history. The ethical dimensions of AI are debated primarily along three axes: (1) the risk of interference with individual and institutional autonomy, (2) privacy violations arising from data collection and processing, and (3) social injustices reproduced through algorithmic biases. These problems require not only technological but also normative, philosophical, and legal inquiry. Moreover, interpretability lies at the center of these ethical debates, because understanding how and why AI systems reach particular outcomes is critical for both user trust and legal accountability. This study examines how ethical issues in AI are intertwined with the interpretability crisis. It offers a multidimensional evaluation of AI’s uses across domains such as healthcare, education, justice, environment, finance, public administration, and social media.