Revisiting the interplay of risks, expected benefits, and ethical implications of AI across varying automation and criticality levels in Arab and UK contexts.

Abstract

Understanding AI's ethical implications is vital for its responsible deployment and societal acceptance. This study revisits the relationship between AI’s potential risks, the positive changes it brings, and its ethical evaluations across cultural contexts, focusing on different levels of automation and criticality. An online survey involving 323 participants from the Arab GCC and 316 from the UK assessed responses to four scenarios, varying by low/high criticality and low/high automation. Hierarchical linear regression examined how perceptions of risk and positive change influence ethical evaluations. Findings show that perceived risks consistently raise ethical concerns, while positive changes mitigate these concerns, especially in high-risk scenarios. Although effects vary between UK and Arab participants, positive changes moderated the relationship between risks and ethical implications in both groups under high criticality. The study provides empirical evidence that perceived risks and benefits jointly influence ethical evaluations across different AI modalities, while cross-cultural variation remains modest.