Machine Learning, Artificial Intelligence, and Algorithm are all terms that exist in the same space of computer science, but they absolutely do NOT all mean the same thing, and in your post here you used them all interchangeably.

An algorithm is a very generic term for some kind of heuristic that can be followed to produce some result. A recipe for cookies in an algorithm just like some algorithm on Facebook decides what posts to show you. Machine Learning takes place when the process a system implements is non-deterministic; it does things that the programmers didn't explicitly tell it to do; it actually learns how to do new things. An artificial intelligence is a system that's designed to do tasks in the same way a human would, often involving processing visual data or making human-like decisions.

If you wanted to make the case that we shouldn't use MACHINE LEARNING in policing, I would 100% agree with that statement, our police policies should be very deliberate and very transparent and machine learning wouldn't be either of those things. But using this as an argument that we shouldn't be embracing policing with explicitly defined algorithms that are far MORE transparent and deliberate than the humans they would replace is an absolutely indefensible argument. If there's one thing we've learned in the past few years, it's that police need far more regulation, and that's exactly what algorithms do whether they are implemented by a computer or by some system of rules and laws.