Return to Article Details Prompt-Induced Bias in LLM Security Recommendations: A Quantitative Behavioral Study Download Download PDF