Generative AI and Translation Ethics: A Context-Sensitive Ethical Protocol for Iranian Translators

سال انتشار: 1405
نوع سند: مقاله ژورنالی
زبان: انگلیسی
مشاهده: 26

فایل این مقاله در 15 صفحه با فرمت PDF قابل دریافت می باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

JR_TIR-3-11_006

تاریخ نمایه سازی: 21 آبان 1405

چکیده مقاله:

Generative AI in translation raises ethical questions about transparency, accountability, privacy, fairness, and human oversight, yet context-specific guidance, particularly in Iran, is limited. This study examined Iranian translation students’ and professional translators’ perceptions of AI ethics and developed a context-sensitive framework and self-assessment checklist. A single-round, Delphi-informed survey of ۴۵ participants (۱۵ undergraduates, ۱۵ MA students, ۱۵ freelance translators) used a ۳۵-item questionnaire covering six ethical dimensions. Descriptive results showed very high agreement on human-AI collaboration and human accountability, and high agreement on disclosure, transparency, privacy, and fairness. Professionals scored lower than students on several dimensions. Drawing on these findings and an existing six-dimensional model, the study proposes a framework of transparency, accountability, human-AI collaboration, privacy, fairness, and ethical governance. Implications for education, practice, and policy include AI literacy, critical oversight, and flexible disclosure guidelines. The checklist translates these principles into practical questions for responsible AI-assisted translation.

نویسندگان

- -

Department of Foreign Languages, Isf.C., Islamic Azad University, Isfahan, Iran

- -

Department of Foreign Languages, Isf.C., Islamic Azad University, Isfahan, Iran