Generative artificial intelligence (AI) has the potential to revolutionize teaching and learning applications. This article examines the word cloud, a toolkit often used to scaffold teaching and learning for reflection, critical thinking, and content learning. Addressing the issues in traditional word clouds, semantic word clouds have been developed but they are technically challenging to develop and still problematic. However, generative AI has the potential to develop efficient, accurate, creative, and accessible word clouds. Three different methods representing three major approaches of word cloud generation were developed, implemented, and user evaluated--traditional (baseline), semantic (natural language processing enhanced), and generative AI (generative pretrained transformer based)--in two different language contexts--Chinese (China case) and English (Singapore case). The findings of the study show the technical robustness of the methods, as well as provide key pedagogical insights from the user perspective of instructors of higher education courses in China and Singapore. Implications to the design of word clouds and their application in teaching and learning are discussed.