Yuritopiaposadism [none/use name] to technology • 10 months agoOpenAI Says It's Fine to Vacuum Up Everyone's Content and Charge for It Without Paying Themexternal-linkmessage-square16 fedilinkarrow-up1101
arrow-up1101external-linkOpenAI Says It's Fine to Vacuum Up Everyone's Content and Charge for It Without Paying ThemYuritopiaposadism [none/use name] to technology • 10 months agomessage-square16 Commentsfedilink
minus-squareJohnBrownNote [comrade/them, des/pair]hexbear23·10 months agoonly if you run it through your own LLM "ai" (and none of this shit is AI, they should have to change their name) "works" aren't copyrightable so go nuts link
minus-squareAwoo [she/her]hexbear3·edit-210 months agoIf ai will regurgitate its training data then you can perform copyright-laundering via this one neat loophole. We can move literally the entire internet (which is basically all in their training data) into the public domain. link
minus-squareJohnBrownNote [comrade/them, des/pair]hexbear3·10 months ago unfortunately i think these things don't keep the training set, just the set of associations and relations it made by analyzing it link
minus-squareAwoo [she/her]hexbear2·10 months agoNot true, they will completely and totally replicate their training data. The companies try to prevent this so the method to get it to happen regularly changes, but they do it. Chatgpt: https://not-just-memorization.github.io/extracting-training-data-from-chatgpt.html Image AIs: https://techcrunch.com/2022/12/13/image-generating-ai-can-copy-and-paste-from-training-data-raising-ip-concerns/?guccounter=1 I'm not saying this would work and you won't get in trouble for doing it. But it would fuck the system just a little bit. link
minus-squareJohnBrownNote [comrade/them, des/pair]hexbear2·10 months agooh wow that 's great lol link
only if you run it through your own LLM"ai" (and none of this shit is AI, they should have to change their name) "works" aren't copyrightable so go nuts
If ai will regurgitate its training data then you can perform copyright-laundering via this one neat loophole.
We can move literally the entire internet (which is basically all in their training data) into the public domain.
unfortunately i think these things don't keep the training set, just the set of associations and relations it made by analyzing it
Not true, they will completely and totally replicate their training data. The companies try to prevent this so the method to get it to happen regularly changes, but they do it.
Chatgpt: https://not-just-memorization.github.io/extracting-training-data-from-chatgpt.html
Image AIs: https://techcrunch.com/2022/12/13/image-generating-ai-can-copy-and-paste-from-training-data-raising-ip-concerns/?guccounter=1
I'm not saying this would work and you won't get in trouble for doing it. But it would fuck the system just a little bit.
oh wow that 's great lol