AI-generated paraphrasing tools, especially QuillBot and Paraphrasing Tool, play a crucial role in preventing plagiarism in academic writing. However, their effectiveness and proficiency have been questioned, particularly regarding the adequacy of their strategies. This qualitative study analyzed the paraphrasing strategies and proficiency levels of QuillBot and Paraphrase Tool. Using a purposive sampling technique, all 30 abstracts from one issue of the "Journal of Second Language Writing" were paraphrased using the two paraphrasing tools in their standard modes, and the results were analyzed using the frameworks of Keck (2014) and Nabhan et al. (2021). The results of the study indicated that both tools primarily used synonym substitution, with QuillBot favoring word-level changes and Paraphrase Tool emphasizing sentence restructuring. QuillBot tended to show minimal revision, followed by moderate revision, while Paraphrase Tool exhibited more moderate revision, followed by minimal and substantial revision. Paraphrase Tool exhibited broader paraphrasing capability than QuillBot, but both tools show some paraphrasing limitations. Overall, while these tools may enhance some writing, writers should thoroughly review the core concepts of the original texts and grammatical structures in specific contexts. For novice writers, paraphrasing practice in classrooms should be conducted under teachers' guidance. AIgenerated tools should be secondary.