Generative AI at Work / Unit 5 GAI 107 short lesson, about 20 seconds
Benchmarks Are Not Your Task
Captions available
The idea
NIST cautions that a model passing tests designed for humans, such as professional licensing exams, does not guarantee it is valid or reliable for that work.
In practice
A tool that scores well on a published legal exam but misreads your organization's own contract templates.
Where people slip
NIST adds that sensitivity to wording and varied real use widen that gap.
Your place on the trail
Every lesson rests on earlier ones and opens later ones. Tap any to jump.
Something wrong in this lesson? Flag it · Sources for Generative AI at Work