A national early warning system powered by artificial intelligence (AI) is steadily transforming how Malaysia supports students on the edge of dropping out. The system analyses each student across seven indicators: attendance, academic achievement, disciplinary record, disability or learning support needs, household income, distance from school, and parents’ marital status. Together, these factors help identify which students may be most at risk, sending automatic alerts to counsellors, teachers and district officials, so they can step in before it’s too late. In 2025, the AI system flagged students at risk of dropout, prompting targeted interventions that enabled more than 9,000 students – of whom 3,287 were girls – to return to school. The system goes beyond identifying students at risk of dropping out. It also asks an important question: once students are re-engaged, how do we keep them in school?