Comment by thatguymike

Comment by thatguymike 5 days ago

Congrats Amsterdam: they funded a worthy and feasible project; put appropriate ethical guardrails in place; iterated scientifically; then didn’t deploy when they couldn’t achieve a result that satisfied their guardrails. We need more of this in the world.

tbrownaw 5 days ago

What were the error rates for the various groups with the old process? Was the new process that included the model actually worse for any group, or was it just uneven in how much better it was?

Reply View 0 replies

jaoane 5 days ago

[flagged]

Reply View 2 replies

nxobject 5 days ago

> because I don't even need to look at the data to know that some groups are more likely to commit fraud.
That is by definition prejudice: bias without evidence. Perhaps they want to avoid that.

Reply View | 1 reply
- jaoane 4 days ago
  
  Thankfully, this project got evidence. Unfortunately, it was shelved.
  
  Reply View | 0 replies