Spremljeno u:
Bibliografski detalji
Glavni autori: Yaulie Deo Y Rindengan, Febrian Rezki Hemeto, Ryan Christian Fabian Rattu
Format: Recurso digital
Jezik:
Izdano: Zenodo 2025
Online pristup:https://doi.org/10.5281/zenodo.17865565
Oznake: Dodaj oznaku
Bez oznaka, Budi prvi tko označuje ovaj zapis!
Sadržaj:
  • <p>ABSTRACT<br>The development of artificial intelligence and machine learning<br>technologies has advanced rapidly over the past decade, yet<br>their implementation presents a range of complex ethical<br>challenges. This study aims to identify and analyze the main<br>ethical challenges in the implementation of artificial intelligence<br>and machine learning through a systematic literature review<br>approach. The research methodology uses the PRISMA protocol<br>with literature searches conducted in the IEEE Xplore, ACM<br>Digital Library, ScienceDirect, Springer Link, and Google Scholar<br>databases for the period 2018–2024. From 287 articles identified,<br>42 high-quality articles were selected for in-depth analysis.<br>The findings indicate that the primary ethical challenges include<br>algorithmic bias and discrimination, data privacy and security,<br>system transparency and explainability, and accountability in<br>automated decision-making. Significant social impacts include<br>job displacement, the digital divide, and information<br>manipulation through deepfakes. The study also identifies<br>various emerging ethical frameworks and risk-mitigation<br>solutions that can be applied.<br>The conclusion highlights that responsible implementation of<br>artificial intelligence requires a holistic approach that<br>incorporates technical, legal, and social aspects, supported by<br>collaboration among developers, regulators, and society.</p>