Escaping Flatland: Understanding Multi-Dimensional Potentials of AI Literacies in College Research and Writing

The concept of “AI literacy” has become popularized in response to the rise of large language models (LLM) and generative AI tools. Faced with a generation of products that are broadly useful but difficult to understand technically, teachers, researchers, managers, and more have propounded “literacy” as a possible solution. Literacy is a powerful concept across domains: it promotes greater accessibility and use, and if AI is becoming more pervasive and powerful, the reasoning goes, we should certainly have more AI literacy so we can be prepared for it. But when we use literacy to signify merely skills competency, we fall into problems that have haunted the concept for decades.