Making plans that specify when and where to perform goal-directed actions—known as implementation intentions—is a powerful self-regulation strategy that promotes goal achievement, particularly when the plans are well-formulated. However, children in particular struggle to formulate plans that meet the criteria to be considered effective. Therefore, the present study investigates whether supportive feedback can help improve the quality of their self-created plans. Specifically, we test the feasibility of employing generative AI to deliver personalized feedback tailored to each child’s input and compare its effectiveness to that of generic, non-personalized feedback created by experts. We plan to recruit 80 children aged 10 to 13 who will participate in a within-subject experimental study. Each child will complete six planning trials in which they first generate a plan, receive either personalized AI feedback or generic feedback, and then revise their plan. We will test whether plan quality improves after AI feedback, and whether AI feedback leads to greater plan quality improvement than high-quality generic feedback. In addition, we examine whether children perceive AI feedback as more helpful and motivating compared to generic feedback. The research findings are expected to (1) advance our understanding of how AI-generated feedback influences children’s planning skills, (2) provide deeper understandings of children’s perceptions on AI-generated output, and (3) inform the design of adaptive feedback systems in education.