With a new academic year and the continuing evolution of AI systems, comes new questions of how to help students and faculty ethically and effectively use AI. Two recent articles inspired me to think about the deskilling effects of AI, the different risk levels of using AI between experts and novices, and the importance of emphasizing that AI is a tool to supplement our human brain power, not supplant it.
A recent study, published in The Lancet, got me thinking about the deskilling effects of AI tools and what this means for students who are in the process of learning skills, specifically legal research skills. The Lancet study found the potential for a deskilling effect from using AI tools in doctors’ abilities to spot cancer. The study was small and there were other limitations, but it adds to existing research into the potential detrimental impacts of using AI and cognitive offloading. For example, Michael Gerlich published “AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking in Societies earlier this year, which highlighted the need for educational strategies that promote critical thinking and critical use of AI tools.
Soon after reading The Lancet article I read Iris van Rooij’s piece on LLRX, “AI Slop and the Destruction of Knowledge”. In her piece, von Rooji describes looking up a definition for “Domain-General Cognitive Ability” and finding an AI generated definition on ScienceDirect that was “AI slop” and a “complete fraud in any conceivable way” (i.e., wrong).