Meta has failed (abysmally) at identifying and categorizing content where you’ve said “show me less of this.”
Bluesky’s not my favorite website but Xblock is proof that the app can go “this is a twitter screenshot and she doesn’t want to see those” at scale.
AI could identify, label, and hide all of these things.
On bluesky it already does: “this is rude” or “this content promotes self harm” , I wish both websites could suppress , snooze, or completely nuke “viral” or political content be it left or right. In bluesky’s case it’s not that I disagree with them. It’s just that I’ve had this shit that I more or less agree with shoved down my throat from every angle for a decade and I’m exhausted and don’t want to see or engage with it anymore. People who have nothing else to say 24/7 every single day of their life and mine just need to go away and I wish the AI on bluesky would just let me filter people whose content is primarily political temper tantrums because I don’t have the time or will to mute or block them all so I just don’t use the product.
In fact for moderation purposes, Facebook already is doing that on their back end. (a few years ago you could see automatically generated alt text like “a woman holding a baby” though I don’t use meta at the present time and don’t know if it’s still doing this.)
AI is already analyzing the memes and purging ones with themes they don’t like on FB though . Unlike bluesky moderation, it’s not presented as something I can leverage or access to make my experience more enjoyable on Facebook.
But that’s not how they’re leveraging AI right now. They won’t let it prevent me from seeing memes posts and content with themes **i** don’t like.
Bluesky’s not my favorite website but Xblock is proof that the app can go “this is a twitter screenshot and she doesn’t want to see those” at scale.
AI could identify, label, and hide all of these things.
On bluesky it already does: “this is rude” or “this content promotes self harm” , I wish both websites could suppress , snooze, or completely nuke “viral” or political content be it left or right. In bluesky’s case it’s not that I disagree with them. It’s just that I’ve had this shit that I more or less agree with shoved down my throat from every angle for a decade and I’m exhausted and don’t want to see or engage with it anymore. People who have nothing else to say 24/7 every single day of their life and mine just need to go away and I wish the AI on bluesky would just let me filter people whose content is primarily political temper tantrums because I don’t have the time or will to mute or block them all so I just don’t use the product.
In fact for moderation purposes, Facebook already is doing that on their back end. (a few years ago you could see automatically generated alt text like “a woman holding a baby” though I don’t use meta at the present time and don’t know if it’s still doing this.)
AI is already analyzing the memes and purging ones with themes they don’t like on FB though . Unlike bluesky moderation, it’s not presented as something I can leverage or access to make my experience more enjoyable on Facebook.
But that’s not how they’re leveraging AI right now. They won’t let it prevent me from seeing memes posts and content with themes **i** don’t like.