AI tools can now derive equations, write solvers and check results. In this two-part discussion we hope to begin a conversation about how we see fluids research changing in light of this rapid AI development, particularly for the theoretical and numerical work common in CFIG. In the first half, Emma will present a case study on her work—heterogeneous flow-induced compaction of soft porous media—to highlight ways in which Claude has changed her workflow and accelerated this research over the past few weeks. This raises pressing questions about how AI-generated work should be checked or credited and gives a starting point for discussing the current capabilities of Claude in low-Reynolds-number theoretical and numerical work. In the second half, Jay will lead a structured discussion on the wider implications of AI for mathematical research, drawing on themes from the Royal Society's *Mathematics in the Age of AI* meeting and other recent initiatives. We will explore how the role of researchers is changing, what this means for publications, research credit, training and careers, and how we can work together to respond to the opportunities and challenges ahead.

Everyone is welcome to join and share their thoughts.