|
|
|
|
|
by delusional
430 days ago
|
|
This is such a radical take on IP rights and AI "learning" that I can only assume you're consciously choosing to misunderstand both. On the off chance that you are not: IP-rights does not cover "learning from" a source. What ML does is not in any way akin to human learning in methodology. When we call it learning that's an analogy. You can not argue a legal case from analogy alone. |
|
Yes they can because the analogy directly applies. The technical details of how a computer learns, vs a human learns doesn't really matter here, and is an irrelevant difference.
The reason why the analogy directly applies is because in both cases it is about how IP cannot really control how someone uses the IP.
Just like how IP laws cannot prevent you from listening to music while standing on your head, it also cannot prevent you from training your models on it (while also standing on your head, lets say!).
Instead, IP laws only prevents the publishing of copies of the IP.
So the point and the analogy stands.