Was any attempt made to remove copyrighted material?
#9
by
umm-maybe
- opened
It's unclear whether any license you attach to this dataset is valid if it contains previously copyrighted text, nor whether commercial use of derived products (such as large language models) is fair.
(I am not a lawyer and this is not legal advice).
It's likely that this dataset does contain copyrighted text.
nor whether commercial use of derived products (such as large language models) is fair.
Whether any use is "fair" in a legal sense is a legal question your lawyer has to answer based on the specific facts at hand and your own jurisdiction. I don't believe the Technology Innovation Institute is going to advise you on that.