As AI models are used by more and more banks for more and more use cases, the validation of such models naturally also becomes a new focus for MRM teams.
Opening up this thread to discuss issues / concerns / tips around the validation of AI models.
What worked well for you? What can be improved?
