Valerie Barr, a computer science professor at Bard College, is voicing concerns about the rapid integration of AI into education, arguing that the current push to incorporate AI into curricula lacks critical examination of its broader consequences. In an article published in the Communications of the ACM, Barr highlights the growing trend of emphasizing AI in teaching without fully addressing the environmental, ethical, and economic implications of this shift. She questions the assumption that every student must engage with AI early and often, suggesting this approach may overlook deeper issues such as the environmental footprint of large language models, data privacy concerns, and the exploitative practices of data workers.
Barr also raises concerns about the over-dependence on a small number of dominant technology companies, which are currently driving AI development. She points out that these firms are burning significant resources in an attempt to achieve even minor performance improvements. However, she warns that this reliance may leave students in a precarious position when these companies eventually plateau or collapse, undermining the educational system’s ability to function without AI assistance. The article underscores a call for a more thoughtful approach to integrating AI into education, focusing on defining the kind of problem solvers and software developers we want to graduate, rather than simply following industry trends.