This research direction focuses on exploring new paradigms for materials discovery enabled by AI Scientist platforms, integrating data-driven modeling with automated experimentation. By establishing a closed-loop research framework that combines high-throughput computation, automated synthesis and characterization, data modeling, and intelligent decision-making, this work aims to synergistically accelerate materials design, synthesis, and validation. Emphasis is placed on developing generative AI and knowledge-driven materials design methodologies, enabling the continuous co-evolution of materials databases and autonomous platforms. This approach significantly enhances research efficiency and supports frontier innovations in energy, catalysis, and related functional materials.