Can Consumers Detect AI Slop? A Scoping Review of Lay Judgments of Flawed Generative AI Health Information

This project examines how consumers evaluate generative AI health information when their judgments can be compared with an independent reference addressing the same content feature. The scoping review maps the characteristics of existing studies, the reference standards used, and the types of flaws consumers recognize or miss, including factual errors, omissions, and unwarranted certainty. It also examines how confidence, source cues, and presentation relate to judgments, verification, trust, and reliance. The conceptual framework integrates dual-process accounts of credibility judgment, signal detection theory, and metacognitive calibration. This repository supports transparent documentation of the review methods and accompanying research materials.