Digital Health Literacy and Ai-Generated Health Information: Its Influence on Health Decision-Making Among Adolescents

Abstract

Generative artificial intelligence (AI) is reshaping the landscape of health information access and adoption for adolescents. Even fluent systems can be used to explain something tricky, but a fluent response can hide unsupported claims, lack context, or may give false confidence. This narrative review aims to explore the potential impact of digital health literacy on adolescents' evaluations of AI-generated health information and their decision-making processes. This draws on literacy instruments, reviews of adolescent online information appraisal and trust, surveys of chatbot use, response-safety evaluations, observational studies, participatory research, and international advice. The literature shows that there is some degree of imbalance in the way people are digitally health literate, but helpful and understandable answers are not necessarily right or safe. Observational evidence shows a correlation between health knowledge acquisition via AI and subsequently physical activity, but fails to confirm causation or directly assess quality of decision. The review indicates that the exposure to comprehension, credibility, independent verification, calibrated trust and action is influenced by the quality of the system, emotional condition, social support, privacy, and access to health care. The existing data does not specify that digital health literacy has a causal protective effect against adolescents who get harmful AI-informed guidance. Responses should demonstrate competency in applying appraisal skills, explain and secure AI systems and provide support that is easily accessible. However, it remains important to measure verification and decision quality prospectively and to separate the literacy intervention effects from improvements in AI design in future studies.