Qorǵau: Evaluating Safety in Kazakh-Russian Bilingual Contexts

Large Language Models (LLMs) are known to have the potential to generate harmful content, posing risks to users. While major progress has been made in developing taxonomies for LLM risks and safety evaluation prompts, most studies have focused on monolingual contexts, primarily in English. However, language-and region-specific risks in bilingual contexts are often overlooked, and core findings can diverge from such in monolingual settings. In this paper, we introduce Qorǵau (meaning 'to protect' in Kazakh), a novel dataset specifically designed for safety evaluation in Kazakh and Russian, reflecting the unique bilingual context in Kazakhstan, where both Kazakh (a low-resource language) and Russian (a highresource language) are spoken. Experiments with both multilingual and language-specific LLMs reveal notable differences in safety performance, emphasizing the need for tailored, region-specific datasets to ensure the responsible and safe deployment of LLMs in countries such as Kazakhstan. Warning: this paper contains example data that may be offensive, harmful, or biased.