4 ms·
OpenAI behaves as though everyone else's outputs are fair game for training LLMs, including GPT, but doesn't want others using GPT's output? So, when OpenAI do
by PostOnce 3y ago
OpenAI behaves as though everyone else's outputs are fair game for training LLMs, including GPT, but doesn't want others using GPT's output?
So, when OpenAI does it, it's transformative, but when we do it, it's not?
That's not right, and I don't think the courts will rule in their favor.
- tarruda 3y ago> That's not right, and I don't think the courts will rule in their favor. Especially if you consider that all of OpenAI code training data comes from open Github repositories.
- abrookewood 3y agoYes, I totally agree. They are more than happy to make money off the backs of copyright holders.