Investigating the Influence of Language on Sycophantic Behavior of Multilingual LLMs

TL;DR

This study examines how the language used influences the tendency of multilingual LLMs to exhibit sycophantic behavior, revealing cultural and linguistic patterns despite progress in mitigation.

Contribution

It provides the first systematic analysis of language effects on sycophancy in state-of-the-art multilingual LLMs across multiple languages.

Findings

Newer models show less overall sycophancy.

Language significantly affects model agreement levels.

Cultural and linguistic patterns influence responses on sensitive topics.

Abstract

Large language models (LLMs) have achieved strong performance across a wide range of tasks, but they are also prone to sycophancy, the tendency to agree with user statements regardless of validity. Previous research has outlined both the extent and the underlying causes of sycophancy in earlier models, such as ChatGPT-3.5 and Davinci. Newer models have since undergone multiple mitigation strategies, yet there remains a critical need to systematically test their behavior. In particular, the effect of language on sycophancy has not been explored. In this work, we investigate how the language influences sycophantic responses. We evaluate three state-of-the-art models, GPT-4o mini, Gemini 1.5 Flash, and Claude 3.5 Haiku, using a set of tweet-like opinion prompts translated into five additional languages: Arabic, Chinese, French, Spanish, and Portuguese. Our results show that although newer models exhibit significantly less sycophancy overall compared to earlier generations, the extent of sycophancy is still influenced by the language. We further provide a granular analysis of how language shapes model agreeableness across sensitive topics, revealing systematic cultural and linguistic patterns. These findings highlight both the progress of mitigation efforts and the need for broader multilingual audits to ensure trustworthy and bias-aware deployment of LLMs.