There is an urgent need for broadening computing education to prepare students for the Artificial Intelligence (AI) workforce, but achieving this vision faces major obstacles. These challenges include the complexity of computational problem-solving, inadequate teacher training, student retention, and the demands of self-paced learning environments where students must develop solutions using AI technologies with limited guidance. Specifically, students often struggle to independently develop problem-solving strategies and effectively manage their learning without consistent support. Self-regulated learning (SRL)—where students actively plan, monitor, and reflect on their own learning process—emerges as a critical skill for success in computing and AI education. While many learners struggle to develop these skills effectively, AI-assisted self-regulation through hybrid intelligence models offers a promising approach. This project leverages hybrid intelligence to develop Meta-Partner, an adaptive AI-assisted SRL solution. Meta-Partner enhances students’ self-regulation and metacognition through close, iterative human-computer collaboration. It empowers students to revise goals, adjust strategies, monitor progress, and enhance self-reflection with continuous AI support throughout the SRL cycle. Meta-Partner will be integrated into AIResolver, an existing online problem-based learning platform for AI literacy. A study with 300 high school and college students, using both quant