Large language models (LLMs)

Definition: 

Machine learning algorithms that can recognize, summarize, translate, predict, and generate human languages on the basis of very large text-based datasets.