As a vital learning activity in second lan-guage (L2) writing classrooms, peer feedback playsa crucial role in improving students' writing skills.However, student reviewers face challenges in provid-ing impactful feedback on peers' essays. Low- qualitypeer reviews emerge as a persistent problem, ad-versely affecting the learning effect of peer feedback.To enhance students' peer feedback provision, thisstudy introduces EvaluMate, an AI-supported peerreview system, which incorporates a chatbot namedEva, designed to evaluate and provide feedbackon student reviewers' comments on peers' essays.Forty-four Chinese undergraduate students engagedwith EvaluMate, utilising its features to generate feed-back on peers' English argumentative essays. Chatlog data capturing the students' interactions with thechatbot were collected, including the comments theywrote on peer essays and the feedback offered by thechatbot on their comments. The results indicate thatthe integration of AI supervision improved the qualityof students' peer reviews. Students employed variousstrategies during their comment revision in responseto AI feedback, such as introducing new points, add-ing details, and providing illustrative examples, whichhelped improve their comment quality. These find-ings shed light on the benefits of AI-supported peerreview systems in empowering students to providemore valuable feedback on peers' written work.