Just like programs only know to do what they're coded to do, algorithms only learn what's already in their dataset. It's up to data scientists and analysts to s

Just like programs only know to do what they're coded to do, algorithms only learn what's already in their dataset. It's up to data scientists and analysts to scrub their dataset of potentially discriminatory data. If you're contractually obligated to do otherwise, you should either find another job or do the right thing anyway. If you choose the latter and someone takes action against you, then you should take it to court. I was in a situation like this in my own work and I regret not standing up for the right thing now...