影響自駕公車決定倫理兩難決策因素之初探

A PRELIMINARY STUDY ON THE DECISION-MAKING FACTORS AFFECTING THE ETHICAL ILEMMA OF SELF-DRIVING BUSES

自駕車是人工智慧的創新應用之一，自駕公車在道路上的行駛實驗也正在世界各地進行中。本研究以Level 4自駕公車為研究對象，建構尖點劇變模型來探討影響自駕公車倫理兩難的決策因素。本研究在臺北與上海以問卷調查收集資料，並且比較兩個城市的調查結果。研究結果顯示有47%的受訪者表示遇到自駕公車倫理兩難時，由「人類司機」或「AI演算法」來進行決策都可接受，且偏好「人類司機」。此外，臺北的受訪者在自駕公車的知覺有用性、使用態度構面有較高的分數，而上海受訪者在知覺風險與知覺行為控制構面有較高的分數。最後，本研究根據所建構的尖點劇變模型進行質性分析，研究結果可提供政府部門研擬自駕公車相關政策時的參考。

Self-driving vehicles are one of the innovative applications of artificial intelligence (AI). On-road experiments with self-driving buses are conducted around the world. This study focuses on Level 4 self-driving buses. We explore decision-making factors of potential users who are faced with a ethical dilemma situation by using a cusp catastrophe model. This study collects data through questionnaire surveys in Taipei and Shanghai. Results of the two cities are compared. We find that 47% of the respondents find it acceptable to allow a ''human driver'' or an ''AI algorithm'' to control the Level 4 self-driving bus when faced with a ethical dilemma. However, they prefer ''human drivers''. In addition, respondents in Taipei perceive a higher usefulness of self-driving buses and have more positive attitudes toward them than respondents in Shanghai. The latter exhibit higher perceptions of risk and behavioral control. In addition, we conduct a qualitative analysis based on a cusp catastrophe model. The results of this study can serve as a reference for government authorities seeking to formulate policies for promoting self-driving buses.