I didn’t plan on writing today, but last night Musk tweeted something right up my alley—about content moderation of “hate speech” on Twitter.
It looks like Musk and his team at Twitter have chosen to algorithmically throttle “hate speech” so it receives fewer impressions. (I’m pretty sure I’m still algorithmically throttled on Twitter for wrongthink, but I digress….)
As you can see, Musk mentioned a list of “bad words”. But he hasn’t answered my question about whether the mere appearance of a “bad word” is enough to get you throttled there. Either way, it will be interesting to see whether throttling this content will be enough to satisfy Google and Apple. (Or other countries’ “hate speech” laws.)
I believe the proper approach is to leave curation of “hate speech” in the hands of the individual user—whether by block/mute, or optional filters—and that’s what we’ve been implementing on Parler.com and sideload Android. But at least when it comes to Parler, Apple and Google have insisted “hate speech” be entirely removed from the apps. Double standard? Time will tell.
For more on “hate” as a category in content moderation, and how we approach this on Parler, see this thread I recently wrote on Twitter:
"Hate Speech" is just the cowardly way of saying, "I don't want to hear that language!" And of course, government has the solution, or in this case a man beholden to the government via SpaceX and other ventures. I'll wait until the dust settles, and then I might give Twitter a try. But is saying I seek my individual rights against all those using force as a primary going to be considered to be "hate speech"?