Tech + discrimination researcher @AmnestyNL.
Likes/shares not endorsements; views my own
Alexander Laufer
ML-based risk profiling systems do not, and cannot perform the social-scientific roles they are saddled with. As such, their use by state actors should be prohibited. Report by @alexlaufer.bsky.social @amnesty.org @amnestynl.bsky.social
Whether predictive systems can be ethically—or even functionally—used in socially sensitive contexts depends on their epistemic credentials. I am delighted to see my work inform policy recommendations in a new report from Amnesty International. www.amnesty.org/en/latest/ne...
When governments use past social data to predict who is going to commit a crime, they inevitably target individuals from marginalized groups and reproduce past injustices and there is simply no technical way to fix such societal issues.
full report: www.amnesty.org/en/documents...
Automatische systemen die door de overheid worden gebruikt om criminaliteit en fraude te "voorspellen", aka risicoprofilering, zijn in strijd met de mensenrechten en moeten worden verboden concludeert @amnestynl.bsky.social in een nieuw rapport
this @amnesty.org report is a must read!!
The belief that machines are neutral produces a façade of objectivity that ends up reinforcing discrimination and inequality. As a result, people suffer from the deployment of tools whose outcomes have been shown to be scientifically dubious & inaccurate. They must be banned!
I feel privileged and immensely grateful to @abeba.blacksky.app, @randomwalker.bsky.social, @bayesianboy.bsky.social, Amanda Coston, Giovanni Sileno, Hilde Weerts, and many others for reviewing this report.
