Generative Language Models in Education: Foreign Language Learning and the Teacher as Prompt Engineer

Generative language models are powerful tools trained on a large digital corpus, and their capability to generate novel text to specification has received significant press for possible interference in education as teachers face submissions which are potentially machine-generated. These problems are acknowledged and elaborated here, but focus is then shifted to how these systems might more productively be inserted into the educational process in the future. The instructional potential of such systems for aspects of foreign language education are explored in an excursive manner, and in the context of arguing that teachers, far from being displaced, will take on a supplementary role as expert user in extracting from these systems precisely what is required for student learning to occur. Such a role is notionally similar to that of “prompt engineer” and identifies an additional and reconfigured space for the teacher as an “educational prompt engineer” of sorts.