Rob Moore, a corporate fellow at Oak Ridge National Laboratory (ORNL), is spearheading a major shift in scientific research through the integration of artificial intelligence and automation. A former U.S. Navy submarine officer turned mechanical engineer, Moore joined ORNL in 2019 to study quantum materials. Recognizing the potential of AI to accelerate the discovery of materials with tailored properties, he initially led the INTERSECT initiative, which established a scalable ecosystem for interdisciplinary self-driving research. Building on this foundation, Moore is now directing the Labs of the Future initiative, aiming to advance ORNL toward fully autonomous operations that support the Department of Energy's Genesis Mission to accelerate discovery science and drive energy innovation. According to Moore, the emergence of large language models has ushered in a new era for scientific inquiry. These tools can process vast amounts of information rapidly, assisting researchers with reliable data retrieval and hypothesis generation. However, Moore emphasizes that reliability and accuracy remain the paramount challenges. Unlike other fields, science cannot afford to propagate errors, as the information produced impacts society and policy. The risk of AI "hallucinations," where models generate plausible but incorrect information, necessitates rigorous validation. Moore stresses that scientific outputs must be accurate, reliable, and reproducible, requiring careful deployment strategies to ensure these tools serve as aids rather than sources of misinformation. AI currently excels at identifying correlations within massive datasets far faster than humans can. It can generate hypotheses and guide experimental directions, effectively offloading cognitive tasks to free researchers for more holistic problem-solving. This capability is particularly crucial for addressing complex grand challenges that have stalled for decades. Moore notes that while the direct application of AI to science was not initially anticipated by the broader community, the rapid development of intelligence in these models has revealed their potential to significantly speed up scientific progress. The distinction between automation and autonomy is central to Moore's vision. Automation involves pre-programmed tasks performed by instruments, such as a 4D Scanning Transmission Electron Microscope using neural networks to identify atomic structures. While efficient, such systems stop after classification without making further decisions. True autonomy requires a decision-making agent that can interpret data and determine the next steps in an experiment. In an autonomous system, the AI does not just execute tasks but actively decides to pursue a new line of inquiry based on its findings, with humans acting in an advisory or oversight role. Moore asserts that while a lab can be automated without AI, achieving true autonomy is impossible without intelligent agents capable of making decisions. This transition from automation to autonomy represents a fundamental transformation in how scientific research is conducted, promising to unlock solutions to previously intractable problems.