xc2215x@lemmy.world to Technology@lemmy.worldEnglish · 2 个月前Digg's new app is basic, but a great startwww.theverge.comexternal-linkmessage-square19linkfedilinkarrow-up115arrow-down166cross-posted to: technology@lemmit.onlinetechnology@lemmy.world
arrow-up1-51arrow-down1external-linkDigg's new app is basic, but a great startwww.theverge.comxc2215x@lemmy.world to Technology@lemmy.worldEnglish · 2 个月前message-square19linkfedilinkcross-posted to: technology@lemmit.onlinetechnology@lemmy.world
minus-squareTetsuolinkfedilinkEnglisharrow-up6·2 个月前I was wondering, is there anything preventing AI to train on the content on Lemmy ?
minus-squareJeena@piefed.jeena.netlinkfedilinkEnglisharrow-up6·2 个月前Difficult, even if your instance blocks it, copies of it are all over the place.
minus-squareMajinBlayze@lemmy.worldlinkfedilinkEnglisharrow-up2·2 个月前If you wanted to train on Lemmy data, just pretend to be an instance and have all the public instances push their data to you. No scraping required, and you get all the metadata and context you could possibly want
minus-squareasudox@lemmy.asudox.devlinkfedilinkEnglisharrow-up2·2 个月前Yep. We have Fediseer for such instances though.
I was wondering, is there anything preventing AI to train on the content on Lemmy ?
No
Difficult, even if your instance blocks it, copies of it are all over the place.
If you wanted to train on Lemmy data, just pretend to be an instance and have all the public instances push their data to you. No scraping required, and you get all the metadata and context you could possibly want
Yep. We have Fediseer for such instances though.