The Trap of AI Literacy: The Paradoxical Relationships Between College Students’ Use of LLMs, AI Literacy, and Fact-checking Behavior

This study examines factors influencing users' critical engagement with large language models (LLMs), focusing on fact-checking behavior.While LLMs transform how individuals acquire knowledge, their rapid adoption raises concerns about uncritical acceptance due to limitations like hallucinations.A survey of college students and young professionals revealed nuanced effects of LLM literacy.Understanding LLM processes, such as input, processing, and output, encourages fact-checking.However, self-efficacy and knowledge of LLM features paradoxically reduce verification by fostering reliance on machine heuristics and elevating the perceived credibility of outputs.These findings highlight the complex role of AI literacy in promoting critical engagement and the importance of education to deepen users' technological understanding.