The ghost in the machine speaks with an American accent: cultural value drift in early GPT-3 and the case for pluralist evaluation of generative AI
Open Access
- 23 March 2026
- journal article
- Published by Springer Nature in AI and Ethics
- Vol. 6 (2)
Abstract
Early large language models (LLMs) were released with minimal alignment, offering a rare view of how generative systems reframe the ethical values embedded in human texts. We examine outputs from a 2021 version of OpenAI’s base GPT-3, prompting it to summarise culturally diverse source materials (laws, political speeches, and philosophical works) and interpreting results through a descriptive, moral value pluralist lens. Where possible, we contextualise outputs with cross-national datasets such as the World Values Survey. We document recurring value drift: Australia’s firearm policy is recast as a threat to liberty; de Beauvoir’s feminist critique becomes gender-essentialist dating advice; and Merkel’s humanitarian appeal is recast as immigration control. In contrast, multilateral documents (UN/UNESCO) exhibit greater value stability, suggesting consensus-crafted language can buffer against cultural mutation. We argue that these early behaviours (observed before extensive fine-tuning and safety layers) provide a baseline for understanding how training distributions shape normative framing. Our contribution is twofold: (1) empirical evidence that value drift can invert or overwrite encoded values along predictable cultural axes, and (2) a pluralist, descriptive evaluation method that surfaces whose values dominate and when. We conclude with implications for culturally inclusive evaluation and alignment in contemporary LLMs.Keywords
All Related Versions
Funding Information
- University of Sydney
This publication has 59 references indexed in Scilit:
- Social Value Alignment in Large Language Models2024
- Word vector embeddings hold social ontological relations capable of reflecting meaningful fairness assessmentsAI & SOCIETY, 2021
- An Introduction to Japanese Society2020
- GPT-3: Its Nature, Scope, Limits, and ConsequencesMinds and Machines, 2020
- Gender Bias in Neural Natural Language Processing2020
- Artificial Intelligence, Values, and AlignmentMinds and Machines, 2020
- The new politics and geographies of scarcityGeoforum, 2019
- Patiency is not a virtue: the design of intelligent systems and systems of ethicsEthics and Information Technology, 2018
- Understanding natural languageCognitive Psychology, 1972
- How to derive ‘ought’ from ‘is’1969