Tuesday, June 28, 2022

Microsoft: Leveraging large language models to build more robust hate speech detection tools. Really!

This is a dangerous trend of language hygiene, an extension of political correctness!

Hate speech is in the ear of the beholder! There is most likely no consensus what exactly constitutes hate speech!

It clearly conflicts with freedom of speech

Censorship usually leads to self censorship or evasion. Suppressed speech may result in violence. 

Microsoft also calls these efforts detoxification! This is the language of Nazis! It reminds of e.g. Eugenics.

(De)ToxiGen: Leveraging large language models to build more robust hate speech detection tools - Microsoft Research

No comments: