It sounds like the best way to bootstrap a machine learning system. You generate the data the system will be seeing in production along with the proper labels. Then in a later stage you can start doing reinforcement learning.
The problem is the lying about it.
That’s sadly pretty on brand.