Rethinking AI Cultural Alignment

TL;DR

The paper challenges the standard one-directional view of AI cultural alignment that relies on static value repositories and argues that cultural alignment emerges from the interaction between humans and AI systems. It proposes a bidirectional framework that queries culturally relevant human values for each AI use case and aligns the AI through interaction designs shaped by users. A GPT-4o case study across four countries demonstrates that alignment levels depend on prompting strategy, with chain-of-thought prompts yielding higher cross-cultural similarity than direct classification, while scenario-based prompts increase unclassifiable outputs. The work highlights the need for context-aware, bidirectional cultural alignment to support culturally diverse and responsible AI deployment, while noting limitations and calling for broader, use-case-driven exploration of value emergence in human-AI ecosystems.

Abstract

As general-purpose artificial intelligence (AI) systems become increasingly integrated with diverse human communities, cultural alignment has emerged as a crucial element in their deployment. Most existing approaches treat cultural alignment as one-directional, embedding predefined cultural values from standardized surveys and repositories into AI systems. To challenge this perspective, we highlight research showing that humans' cultural values must be understood within the context of specific AI systems. We then use a GPT-4o case study to demonstrate that AI systems' cultural alignment depends on how humans structure their interactions with the system. Drawing on these findings, we argue that cultural alignment should be reframed as a bidirectional process: rather than merely imposing standardized values on AIs, we should query the human cultural values most relevant to each AI-based system and align it to these values through interaction frameworks shaped by human users.