Apr 27
Representational Harms in LLM-Generated Narratives Against Global Majority Nationalities
★★★★★
significance 3/5
This research paper examines how large language models exhibit representational harms and biases against Global Majority nationalities. The study finds that LLMs frequently produce harmful stereotypes and one-dimensional portrayals of certain national identities, particularly when US-centric cues are present in prompts.
Why it matters
Systemic biases in LLM-generated narratives threaten to institutionalize Western-centric stereotypes against the Global Majority in automated content generation.
Tags
#llm bias #representational harm #cultural bias #social impactRelated coverage
- Global South OpportunitiesPivotal Research Fellowship 2026 (Q3): AI Safety Research Opportunity - Global South Opportunities
- arXiv cs.AIAn Intelligent Fault Diagnosis Method for General Aviation Aircraft Based on Multi-Fidelity Digital Twin and FMEA Knowledge Enhancement
- arXiv cs.AIPExA: Parallel Exploration Agent for Complex Text-to-SQL
- arXiv cs.AIThe Power of Power Law: Asymmetry Enables Compositional Reasoning
- arXiv cs.AIOn the Existence of an Inverse Solution for Preference-Based Reductions in Argumentation