The massexodus from X really gather steam this weekwith the Bluesky alternative surpassing 17 million users in a stir of anti Elon Musk sentiment .
However , while meg are embracing a new social medium world less skew less by algorithmic rule surfacing more and more correct wing ideology , not all of us have dug into whether Bluesky has the potential to be any better , or eventually flow dupe to the same pitfalls .
Well , we just catch one of those answers . Unlike X , Bluesky hasvowed not to train generative AIby grate the depicted object posted to the political program by users .
In a post from the official explanation today , Bluesky clarified its generative AI policies and swan its commitment to writers and creative person who ’ve find a newfangled billet to post over the last duet of weeks .
“ A issue of artists and Maker have made their home on Bluesky , and we hear their concerns with other platform training on their data point . We do not use any of your content to check procreative AI , and have no intention of doing so . ”
The company said that , although some AI is used in recommend content and for inner depicted object moderateness , none of those are Gen AI systems that swear on drug user inputted content for training .
The screw thread continues : “ Bluesky apply AI internally to assist in subject matter moderation , which facilitate us triage spot and harbor human moderators from harmful mental object . We also use AI in the Discover algorithmic provender to serve you posts that we think you ’d like . None of these are Gen AI systems train on user content . ”
Back in July this year it emerged a default setting on X has been giving the site permit to train its Grok AI tool using your posts and fundamental interaction , without you ever enjoin yes explicitly . gratefully , users were able to plough it off , but had to manually do so . If you ’re concerned here’show to discontinue Elon Musk develop his Grok AI on your tweet .
