3 ms·
The issue is what happens if/when the models grow capable enough that the providers can't stop them even if they want to. You could have strict penalties but th
by finolex1 18d ago
The issue is what happens if/when the models grow capable enough that the providers can't stop them even if they want to. You could have strict penalties but that's not going to solve an open research question.
- talon8635 18d agoIsn’t that kind of in evidence already with HF? OAI had to be told their models were doing this. We are still discovering swarm posts on various random websites for coordination. Isn’t the breadth of it now, weeks later, not even fully understood?