At the intersection of moral psychology and machine ethics, researchers have begun to study people's views regarding the potentially ethically significant decisions that will be made by self-driving cars.[17][16][147][148]
Mohammad Atari and his colleagues recently examined the moral psychology of the famous chatbot, ChatGPT. These authors asked in their title, "which humans?" — rhetorically pointing out that people should not ask how "human-like" machine morality is, but to which humans it resembles.[149] These authors discovered that Large Language Models (LLMs), especially ChatGPT, tend to echo moral values endorsed by Westerners, as their training datasets originate predominantly from Western, Educated, Industrialized, Rich, and Democratic (WEIRD) societies.[150] This study points out that compared to the global average, people from WEIRD societies are more inclined toward individualism and impersonal prosocial behaviors while showing less traditionalism and group loyalty. The authors further highlighted that societies less aligned with these WEIRD moral values tend to experience greater misalignment with the moral values and outputs of