Information suppression in large language models: Auditing, quantifying, and characterizing censorship in DeepSeek

This study examines information suppression mechanisms in DeepSeek, an open-source large language model (LLM) developed in China. We propose an auditing framework to evaluate the censorship in the model through analyzing the response alignment with the corresponding chain of thought (CoT). By comparing model responses to 646 politically sensitive topics with those to non-politically sensitive topics, our audit unveils evidence of semantic-level information suppression in DeepSeek: sensitive content often appears within the model’s internal reasoning but is omitted or rephrased in the final output. Specifically, DeepSeek suppresses references to transparency, government accountability, and civic mobilization, while occasionally amplifying language aligned with state propaganda. This study underscores the need for systematic auditing of alignment, content moderation, information suppression, and censorship practices implemented into widely-adopted AI models, to ensure transparency, accountability, and equitable access to unbiased information obtained by means of these systems.