3. So question is never “individualized judicial sentences vs. statistical algos.” It’s just WHO does a BETTER job making statistical calls.
-
-
Replying to @JohnFPfaff
4A. Maybe it’s judges, maybe it’s algos. Not the point here (though history in psych at least is… pretty clear). Just need a better debate.
3 replies 0 retweets 0 likes -
Replying to @JohnFPfaff
4B. Judges and algos make same sort of population-based statistical decisions, even if judges think otherwise. Algos just more transparent.
1 reply 1 retweet 0 likes -
Replying to @JohnFPfaff
5. So debate has to be over which has less bias now, which can be trained/programmed to have less bias going forward. (Bet algos win #2.)
1 reply 0 retweets 0 likes -
Replying to @JohnFPfaff
6. That’s what made ProPublica article so frustrating: gave error rates for model, not for judges. Impossible to make comparative assessment
1 reply 0 retweets 2 likes -
Replying to @JohnFPfaff
7. But proprietary point is big. Algos used by the govt should not be secret. Esp. at this early stage, need a good public debate abt them.
2 replies 0 retweets 3 likes -
Replying to @JohnFPfaff
That human judgment has issues is true; but at least widely understood. Here, computation is presented as objective, neutral.
2 replies 0 retweets 0 likes -
Replying to @zeynep @JohnFPfaff
Computation has role as a tool to balance out human judgement—but this example shows its fetishization may make things worse.
2 replies 0 retweets 2 likes -
Replying to @zeynep @JohnFPfaff
Go from world of human judgment that we understand has huge problems to computerization that people treat like infallible magic?
1 reply 0 retweets 0 likes
I dunno; they are being implemented in more states; and the world is moving in that direction. Must start out right.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.