no, I did read up on synthetic datasets a bit a while ago, when I learned that robots are being trained to navigate the world in video games - where all that additional data is already available and doesn't need to be annotated.
Plus, a whole bunch of other data that's not properly human-intelligible, and how that created things like instagram face filters and such.
But I know nothing about actual human tasks in machine learning.
_________________
I can read facial expressions. I did the test.