Year: 2026
Type: article
Abstract: Background Generative artificial intelligence (AI) is increasingly embedded in the contemporary health information environment, where it can both generate and mitigate misinformation. Objective This review develops a Lifecycle-Based Framework to provide an integrative account of how generative AI operates across the health misinformation lifecycle from misinformation production and detection to correction and impact. Methods A narrative integrative review was conducted, drawing on research from communication science, public health, human-computer interaction, and artificial intelligence. Results The review identifies domain-specific mechanisms through which generative AI operates across the health misinformation lifecycle. In production, AI enables scalable generation of fluent, expert-styled content through mechanisms including hallucination, amplification of falsehoods, adversarial vulnerability, and deepfakes. In detection, it supports automated fact-checking, pattern recognition, and classification, while remaining vulnerable to false positives, false negatives, bias, and adversarial adaptation. In correction, generative AI facilitates rebuttals, tailored messaging, conversational engagement, and AI-assisted inoculation, but also raises concerns about trust displacement, epistemic authority, and accountability. At the impact stage, AI-mediated misinformation can influence individual cognition and behavior as well as broader social and institutional outcomes, with fluency, narrative coherence, and personalization serving as important persuasive mechanisms. The framework further conceptualizes the lifecycle as recursive and adaptive, such that production, detection, correction, and impact continually shape one another. Conclusion Generative AI should be understood not as a single source of misinformation risk, but as a multifunctional component of a dynamic health communication system. Addressing AI-mediated health misinformation therefore requires coordinated, multilevel strategies that account for its simultaneous roles in producing, detecting, correcting, and shaping the consequences of misinformation, while preserving human oversight, institutional accountability, and public trust.
Source: Frontiers in Medicine
Author: Xiaoli Nan
Institution: University of Maryland, College Park
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7223294351