-against our own biases- we are evaluating and validating models all the time. There are many, many models with acceptable accuracy that we rely on every day.

-against our own biases- we are evaluating and validating models all the time. There are many, many models with acceptable accuracy that we rely on every day. I think this is a myopic view of the current state of ML.