Will AI narrow science's paradigms?

How about one more from Asimov before they go: this week Alvin Djajadikerta published “Designing AI for Disruptive Science.” In it, he explores the possibility that (current approaches to) AI-for-science might enhance our ability to predict within current frameworks while simultaneously weakening our capacity to identify and shift into entirely new paradigms. He lays out histories of important paradigm shifts, discusses some interesting papers on technology-based scientific narrowing (arising from both AI and, interestingly, the internet), and closes with a call for more and better study of the mechanisms of science – perhaps using AI agents as a metascientific model organism for better understanding the effects of different institutional and incentive designs (I, for one, would immediately subscribe to a science-focused AI village-style experiment). It’s a thoughtful and thought provoking piece, and well worth the read.