Autonomous catalysis research with human–AI–robot collaboration

Abstract

Catalysis is essential to modern chemical manufacturing and environmental sustainability. Yet, traditional catalyst discovery remains slow, resource-intensive and constrained by human-centred trial-and-error workflows. The integration of artificial intelligence (AI), robotics and high-throughput experimentation into self-driving laboratories (SDLs) presents a transformative approach for accelerating catalyst discovery and optimization. SDLs combine automated synthesis and testing platforms, data infrastructures and AI-guided decision-making to enable information-rich experimentation and the fast-tracked generation of scientific knowledge. However, in our view, realizing the full potential of SDLs requires sustained human oversight to ensure rigorous data curation, validate machine-generated hypotheses and establish benchmarks to mitigate AI-related errors. This Perspective outlines core SDL components, including hardware, computational modelling and AI-guided decision-making. We discuss challenges in data availability, integration of computational and experimental workflows and scalable platforms. Finally, we outline immediate opportunities to broaden the adoption of autonomous experimentation in catalysis.