[belter]

I understand and appreciate your clarification. However would it not be the case some synthetic data strategies, if misapplied, can resemble the feedback loop scenario and thus risk model collapse?