AI in science - which field gets transformed first?

AI is transforming science unevenly. Which field do you think gets transformed first, and what is the bottleneck there?

Marking this - our project starts something similar next month, learning from everyone's experience.