Businesses and organisations often include environmental claims in their product advertisements, but some can be misleading or false. This practice is colloquially known as “greenwashing”, which can erode consumer trust and hinder genuine efforts towards sustainability. To maintain a trustworthy advertising environment, it is crucial to systematically identify and reveal said misleading advertisements. However, with the significant increase in the number of advertisements being served digitally, manual screening of online platforms becomes time-consuming and inconsistent. Therefore, we present an automated system to identify advertisements containing environmentally-friendly claims (i.e., green claims). Unlike previously established models that rely on neural networks, we propose a logistic regression model because it offers improved explainability. Specifically, each word has a coefficient representing how much it influences the model’s prediction, and each prediction is associated with its uncertainty to assist further scrutiny. We also compare the features of green advertising between Australia and the US, showing a significant difference in their terminology. The findings demonstrate the potential of machine learning and explainable AI (XAI) in addressing greenwashing and promoting more effective and trustworthy green marketing practices, ultimately fostering a healthier advertising ecosystem.