<![CDATA[The increasing use of online games among vocational high school students has raised concerns regarding students’ digital habits and the need for more meaningful technology-supported learning experiences. This study aimed to develop the SWITCH-EL Model (Switching Digital Gaming Engagement into English Learning) by integrating AI-generated animated videos, interactive speaking activities, learner-generated digital content, and reflective digital practices to support English speaking development and digital self-regulation. The study employed a Design-Based Research (DBR) approach with a mixed-method evaluation. A preliminary needs analysis involving 300 vocational high school students in Pekanbaru, Indonesia, identified substantial online gaming engagement, with 45.67% of students reporting more than six hours of gaming per day. The SWITCH-EL Model was developed through six stages: Self-Monitoring Digital Habits, Watch AI-Animated English Content, Interactive English Challenge, Talk and Task Performance, Create AI-Based English Content, and Habit Reflection and Time Rebalancing. The results showed a marked descriptive improvement in English speaking performance, with the proportion of students achieving scores of 75 or above increasing from 9.33% in the pre-test to 81.67% in the post-test. Qualitative interviews further indicated increased awareness of digital habits, better management of gaming and learning time, greater engagement in English learning, and more purposeful use of digital technology. These findings suggest that SWITCH-EL has potential as an innovative instructional model for integrating AI-supported language learning with more responsible digital engagement among vocational high school students]]>
