Distillation
Distillation trains a smaller "student" model to imitate a larger "teacher" model's answers. The student is cheaper and faster, keeping much of the teacher's skill for common tasks.
In one line, for a 12-year-old
Distillation is a big AI tutoring a smaller AI until the little one is almost as good.
An example
A company offers a "mini" or "flash" version of its model that was distilled from its flagship.
Why it matters to people
Cheaper models put AI within reach of more people and organisations. Distilling a rival's model without permission, though, is contested and may breach its terms.