"A language model that has been trained on such data will pick up these kinds of problematic associations."
"Such data" is basically everyday, normal discourse, and some of the "problematic associations" are training that includes phrases like "woman doctor", "both genders", etc. While I get the point, this itself is a biased interpretation of discourse and would be worrisome imho to have people filtering models with their own biases vs the language as it's used by the vast majority of people.
And that's not even getting into the whole "global warming is racism" digression. Maybe it is, maybe it isn't, but it's definitely not a topic for an AI paper. Just say carbon is bad!
It seems like the goal of the paper is to make a bunch of sociological statements and impute the authority of "AI Ethics Lead Researcher" to them -- they're just political opinions, everyone has those.
[1] Values of Machine Learning, William Agnew, Abeba Birhane, Dallas Card, Ravit Dotan, Pratyusha Kalluri at Resistance AI Workshop 2020, https://drive.google.com/file/d/1tjrm3Bf1hxV8iuPSiCcM1IazITG...,
