Found the same story published 3 years ago with better title: "Study finds racial bias in tweets flagged as hate speech"

https://as.cornell.edu/news/study-finds-racial-bias-tweets-f...

>“When we as researchers, or the people we pay online to do crowdsourced annotation, look at these tweets and have to decide, ‘Is this hateful or not hateful?’ we may see language written in what linguists consider African American English and be more likely to think that it’s something that is offensive due to our own internal biases,” Davidson said. “We want people annotating data to be aware of the nuances of online speech and to be very careful in what they’re considering hate speech.”

The actual paper says more:

https://arxiv.org/pdf/1905.12516.pdf

> many of the AAE tweets also use words like “n*gga” and “b*tch”, and are thus frequently associated with the hate speech and offensive classes

As a minority myself I’ve witnessed this more times than not

The original paper (https://arxiv.org/pdf/1905.12516.pdf) draws a different conclusion from what the article claims. Read Section 5 (Discussion) for details. Also, the analysis seems to be limited to looking at just two words, and to two groups (black and white), so I'm not sure how this generalizes. (the latter also makes the title of the post incorrect)

The OP post is also a right-wing-nonsense website that can't even get the affiliation of the authors correct ("University of Cornell").

Mods, please replace the link with a link to the paper, or to a more reputable academic source.

Aren't minorities the only groups "allowed" to use the derogatory words for their own group? Seems like this finding follows.

Once again demonstrating that this puritanical push to purge all “bias” from machine learning models is a fool’s errand, and a transparently ideological one at that.

Anyone who attends a diverse public school knows this. People take pride where they’re from.

Maybe its subtlety. Not all thought crime is keyword oriented.

Hate speech kind of is…