fig1

From large language models to AI agents in energy materials research: enabling discovery, design, and automation

Figure 1. Evolution from LLMs to AI Agents in Energy Materials Research. Phase 1 depicts large language models (LLMs) acting as knowledge-augmentation tools, where human researchers employ prompt-based interaction for data extraction, literature mining, and property prediction, functioning as a “Knowledge Co-pilot”. Phase 2 illustrates an AI-driven human-robot collaboration, in which an AI Agent with an LLM-driven core autonomously plans, executes, and analyzes experiments. Robotic systems carry out experimental tasks under human supervision, closing the loop from hypothesis generation to data validation. AI: Artificial intelligence.