Michael Ten @lemmy.world to Technology@lemmy.worldEnglish · 2 years agoOpenAI transcribed over a million hours of YouTube videos to train GPT-4www.theverge.comexternal-linkmessage-square43linkfedilinkarrow-up1167arrow-down112
arrow-up1155arrow-down1external-linkOpenAI transcribed over a million hours of YouTube videos to train GPT-4www.theverge.comMichael Ten @lemmy.world to Technology@lemmy.worldEnglish · 2 years agomessage-square43linkfedilink
minus-squareGrimy@lemmy.worldlinkfedilinkEnglisharrow-up12arrow-down1·2 years agoIf they didn’t, google would own all video data just like Reddit would own most conversation data. It’s a good thing, at least this lets the open source scene be a player.
minus-squareEccitaze@yiffit.netlinkfedilinkEnglisharrow-up6arrow-down5·2 years agoIn what world is OpenAI open source?
minus-squareGrimy@lemmy.worldlinkfedilinkEnglisharrow-up3·2 years agoThey aren’t but if they can disregard ToS, so can the open source scene.
If they didn’t, google would own all video data just like Reddit would own most conversation data.
It’s a good thing, at least this lets the open source scene be a player.
In what world is OpenAI open source?
They aren’t but if they can disregard ToS, so can the open source scene.