The study aims to assess the effectiveness of ChatGPT in improving grade XI students' writing skills at SMAN 1 Puloampel. The method used was a quantitative approach with a quasi-experimental design involving 60 students divided into a control group and an experimental group. Data were collected through pre-tests and post-tests in the form of essays. The results of the analysis using the Independent T-test showed a significance value (2-tailed) of 0.000, which means that the null hypothesis (H0) can be rejected, supporting the alternative hypothesis (Ha). Moreover, the effect size test yielded a value of 1.18, indicating a "high effect" category according to Cohen's interpretation. It suggests that ChatGPT was an effective medium in improving the writing skills of eleventh-grade students at SMAN 1 Puloampel, particularly in explanation text material.