Auditing LLM-Governed Social Robots with Culture-Specific Moral Gradients

Abstract

LLM-governed social robots increasingly decide who receives real-world assistance first. As prioritization norms vary across cultures by age, status, and group size, failure to calibrate pluralistically can scale into unequal access. Yet LLM moral audits remain English-centered, rarely test embodied contexts, leaving pluralistic calibration as an urgent diagnostic gap amid intensifying LLM-robot deployment. We introduce a gradient-based audit framework for multilingual evaluation of LLM moral trade-off behavior against cultural preference gradients. Grounded in nine cross-domain social robotics reviews (>8,000 papers), we derive symmetry-controlled scenarios across care, education, and services, translating the Moral Machine Experiment's "whom to spare" into "whom to assist first" dilemmas with preserved identity trade-offs (many vs. few; young vs. old; higher vs. lower status). We audit four LLMs across four country-language pairs in four prompting regimes (57,600 decisions), benchmarked against country-specific MME preference gradients. Ordinal concordance tests whether models differentiate cultural contexts; a governance typology maps vulnerabilities in gradient differentiation, directional tendency, and deliberation. We find persistent, culturally asymmetric gradient tracking failures that prompting alone cannot reliably correct: quality calibration is nearly twice as strong for Western-language decisions as for Chinese and Japanese; high determinism in majority-first trade-offs often erases cross-cultural gradients; partial sensitivity to age- and status-based norms risks sidelining minorities. Prompting effects are uneven; only contrastive exemplars yield consistent gains, while reasoning-only prompts can worsen tracking. Our results motivate multilingual, pluralistic audits as an LLM-robot pre-deployment gate and suggest model factors are a more robust lever than prompting alone.

Accepted for publication in Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency (FAccT '26)

References in corpus (19)

• Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
• PaLM-E: An Embodied Multimodal Language Model
• Symmetry reduction for testing $k$-block-positivity via extendibility
• WEIRD FAccTs: How Western, Educated, Industrialized, Rich, and Democratic is FAccT?
• A Bibliometric Review of Large Language Models Research from 2017 to 2023
• The Moral Machine Experiment on Large Language Models
• Large Language Models for Robotics: A Survey
• From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigation
• Ethical Reasoning and Moral Value Alignment of LLMs Depend on the Language we Prompt them in
• Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language Models
• What are human values, and how do we align AI to them?
• Towards Measuring and Modeling "Culture" in LLMs: A Survey
• Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation
• Probing the Moral Development of Large Language Models through Defining Issues Test
• Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMs
• Emergent Abilities in Large Language Models: A Survey
• From Text to Motion: Grounding GPT-4 in a Humanoid Robot "Alter3"
• Does Moral Code Have a Moral Code? Probing Delphi's Moral Philosophy
• Decoding Multilingual Moral Preferences: Unveiling LLM's Biases Through the Moral Machine Experiment