Studies consumer perceptions of AI in response to minor versus severe errors and whether these involved technical failures of the algorithm or violations of social norms. Examines whether explainable AI (XAI) can mitigate the effects of these different types of errors.