HMN 2026: How LLMs stereotype non-Western moral values in predictable ways,

Aliah Zewail, a graduate student in psychological and brain sciences in the College of Natural Sciences (CNS), has led research for a new paper examining the confluence of artificial intelligence (AI), large language models (LLMs), morality, and cultural diversity.

In “Moral Stereotyping in Large Language Models,” published in the Proceedings of the National Academy of Sciences, Zewail and her co-authors detail their findings that LLMs “stereotype the moral values of non-Western populations in predictable ways.”

Lead author Zewail wrote the paper in partnership with corresponding author Mohammad Atari, assistant professor of psychological and brain sciences in CNS, as well as Alexandra Figueroa of the University of California, Berkeley, and Jesse Graham of the University of Utah.

“We provide a cautionary tale of researchers using LLMs as generators for cultural data,” Zewail says of the paper. “Specifically, we provide insight into the value structure of LLMs and how they misrepresent the moral profiles of less-WEIRD (Western-educated, industrialized, rich, and democratic) nations, portraying them as less morally concerned than they actually are.”

The paper goes on to argue that generative pre-trained transformers (GPTs)—such as ChatGPT—play a role in this misrepresentation of the moral values of non-Western individuals.

“We showcase that social scientists should not replace human participants with LLMs, as these AI systems fail to capture human diversity around the globe,” Zewail asserts.

The paper comes on the heels of Zewail receiving supprt for her research on moral values and the cultural evolution of democratic norms in the Middle East and North Africa.

“The application of this paper is critical as AI systems increasingly assume more involvement in some researcher (and laypeople) decision-making processes,” Zewail contends.

“By addressing potential inaccuracies in a model’s output, their work aims to inform policy that prompts ethical AI development and usage to enhance the reliability, accuracy, and cultural competence of what is becoming a ‘global’ tool.”

The content is provided for information purposes only.