3 ms·
If they open source it, everyone would know that they used fuck ton of pirated content to train their models
by udev4096 3y ago
If they open source it, everyone would know that they used fuck ton of pirated content to train their models
- andybak 3y agoAs far as I'm aware training does not currently constitute "piracy". It's fine to advocate for a redefinition but be explicit about it.
- gnomewascool 3y agoI think the point here is about the procurement of the training data, in violation of copyright laws ("piracy"), rather than that the training itself is piracy. The suspicion[0] is that OpenAI trained their models on a large text dump including libgen (in the so-called "books2"). If a person downloads a book from Library Genesis, they're a pirate; if OpenAI does it, so are they. [0] https://twitter.com/theshawwn/status/1320282152689336320 https://twitter.com/theshawwn/status/1320282152689336320