Since the late 1990s, food safety certification has emerged as a prominent and influential regulatory mechanism in both the private and public spheres of the contemporary agri-food system. Food safety standards protect consumers from foodborne illnesses and help producers avoid the massive economic losses associated with food safety breaches. We empirically examine the relationship between foodborne disease outbreaks and certification adoption by utilizing the data on food safety certification adoption in the United States and Europe from 2015 through 2020. In our regression models, food safety certification along with select economic variables such as gross domestic product are used to explain the number of illnesses caused by foodborne disease outbreaks. For the United States at the state level, we found that certifications to SQF, PrimusGFS, BRC, or FSSC 22000 are negatively associated with the number of foodborne illnesses. For the case of Europe at the country level, certifications to ISO 22000 or FSSC 22000 are negatively associated with the number of foodborne illnesses. We then proceed to use machine learning techniques to examine how well we can use food safety certification data to predict foodborne disease outbreaks. Applying several algorithms (ordinary least squares, multinomial, decision tree, and random forest) to the U.S. data, we found that our models with food safety certification adoption can predict the number of U.S. foodborne illnesses or deaths with a relatively high degree of precision (testing accuracy at around 70% or better). Feature importance analysis allows us to inspect the relative importance of each explanatory variable (or feature) for making accurate predictions of the illness or death numbers. Through ranking the importance of explanatory variables, our study reveals that certification information could be the second most important variable (after gross domestic product) contributing to explain foodborne disease outbreaks.