Ethics and Bias in LLMs: Social Responsibility in Generative Artificial Intelligence
DOI:
https://doi.org/10.52072/unitek.v18i1.1386Keywords:
Large Language Models, bias, etika, tanggung jawab sosial, narrative literature review.Abstract
Large Language Models (LLMs), as part of generative artificial intelligence, have brought significant advancements in natural language processing. However, numerous studies indicate that LLMs are not free from bias and potential discrimination, whether in training data or model design. This study aims to analyze the ethical and social responsibility issues in the development and deployment of LLMs, focusing on bias and its implications for social justice. The method employed is a narrative literature review of academic literature, policy reports, and relevant industry documents. Findings reveal that bias in LLMs is systemic and multidimensional, and current governance mechanisms are inadequate. This research recommends transparency, ethical audits, multi-stakeholder engagement, and participatory research approaches as mitigation strategies. Hence, future LLM development must prioritize not only technological efficiency but also justice, accountability, and inclusiveness..
Downloads
References
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? FAccT 2021 - Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610–623. https://doi.org/10.1145/3442188.3445922
Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference, 1(Mlm), 4171–4186.
Floridi, L., & Cowls, J. (2022). A unified framework of five principles for AI in society. Machine Learning and the City: Applications in Architecture and Urban Design, July, 535–545. https://doi.org/10.1002/9781119815075.ch45
Jasanoff, S. (2004). States of knowledge: The co-production of science and the social order. In States of Knowledge: The Co-Production of Science and the Social Order. https://doi.org/10.4324/9780203413845
Mitchell, M., Wu, S., Zaldivar, A., Barnes, P., Vasserman, L., Hutchinson, B., Spitzer, E., Raji, I. D., & Gebru, T. (2019). Model cards for model reporting. FAT* 2019 - Proceedings of the 2019 Conference on Fairness, Accountability, and Transparency, Figure 2, 220–229. https://doi.org/10.1145/3287560.3287596
Sari, F., & Mahmud, S. F. (2024). Implementasi Metode Preference Selection Index Dalam Memilih Media Pembelajaran Matematika Barbasis Artificial Intelligence. Jurnal Unitek, 17(1), 141–151. https://doi.org/10.52072/unitek.v17i1.869
Stilgoe, J., Owen, R., & Macnaghten, P. (2013). Developing a framework for responsible innovation. Research Policy, 42(9), 1568–1580. https://doi.org/10.1016/j.respol.2013.05.008
