It doesn't work well enough yet. The flashcards it generates don't actually fit well into its own ecosystem. When you try to build the "quizzes", the wrong answers are trivially spottable. Further, even the generated questions are stilted don't hit parity with manually generated flashcards.

My use of ChatGPT for this purpose is so far mostly limited to a sanity check, e.g. "Do these notes cover the major points of this topic?" Usually it'll spit back out "Yep looks good" or some major missed point, like The Pacific Railway Act of 1862 for a topic on the Civil War's economic complexity.

I'll also use it to reformat content, "Convert these questions and answers into Anki format."