I think there are valid reasons to not trust an algorithm to determine whether someone is a flight risk or danger to the community. But there was already an a

I think there are valid reasons to not trust an algorithm to determine whether someone is a flight risk or danger to the community. But there was already an algorithm that came up with the cash bail amount for a defendant, which was purportedly based on the defendant's flight risk and likelihood of endangering the communoty. I don't pretend to be an expert so I may be wrong but I think the algorithm boiled down to something like "let the judge decide". I think transparency around the algorithm is incredibly important, so anything that explicitly makes the properties of the algorithm part of the debate I think is good.