AI for Science: Reframing AI’s Role in Discovery
TL;DR
This paper explores how AI contributes to science, introducing a framework organised around task capabilities, scientific workflow integration, and domain constraints, and uses that framework to open wider questions about the role of AI in scientific discovery.
Abstract
Digital computers have reshaped scientific practice, moving working scientific knowledge from printed texts into algorithms, simulations, and models. Advances in artificial intelligence (AI) are now accelerating that shift, progressing science in areas from protein folding to climate modelling, and raising the prospect of a further transformation in how science is done. With growing hype around the field, there is a risk that inflated claims about AI’s potential obscure both its current limitations and its longer-term possibilities. This paper explores how AI contributes to science, introducing a framework organised around task capabilities, scientific workflow integration, and domain constraints. It uses that framework to open wider questions about the role of AI in scientific discovery. These include: Is scientific knowledge constructed and used by AI agents considered scientific understanding if it is impenetrable to humans, or does scientific understanding refer to an activity that is intrinsically human? What technical advances are needed to move AI beyond pattern matching toward causal reasoning? And what institutional changes are needed to support responsible AI adoption? How researchers and policymakers engage with these questions will shape whether AI accelerates progress within existing scientific paradigms or catalyses the generation of new forms of scientific knowledge. This paper marks the opening of a call for papers from RSS Data Science and AI, which invites contributions that take up these and related questions from multiple perspectives.
