4 ms·
It's not exactly practical to hand out the training material given the sheer quantity of data we're talking about.
by dralley 2y ago
It's not exactly practical to hand out the training material given the sheer quantity of data we're talking about.
- philipkglass 2y agoBut they could provide training code and let people provide their own Common Crawl (or whatever other pile of training data), couldn't they?
- stonogo 2y agoYeah, no. We can move an arbitrary amount of data around the world at breakneck speed. Netflix does this for a living. It's not practical to hand out the training material because of the massive rampant copyright violations.
- llm_trw 2y agoIf only we'd figured out a technology that let us move huge torrents of bits around. If only there was a catchy name for it. Something like bit-torrent perhaps?
- mossTechnician 2y agoIf a research group downloads material in order to train a model, is there some significant difference in copyright violation if they hand it to a second research group in order to fulfill the same purposes?
- Onawa 2y agoYes, because of a key word in a lot of copyright laws... "distribution". Using that copyrighted material themselves to train the model still gives them plausible deniability. Handing the copyrighted material to another group starts to run afoul of other laws and also removes the plausible deniability that the original group can claim regarding their training data.
- TZubiri 2y agoThe training data is not necessarily kept. It's possible that data is consumed, incorporated into the weights and then discarded.
- andrewf 2y agoGPL v2 and earlier let you charge distribution costs (v3's language is more complicated). In the late 80s you could order an Emacs tape from the FSF for $150, which is about $430 today!