[OT] Why governments must limit AI violations of copyright

Mary Spender presents a relatively brief but, I think, compelling argument for why governments need to reject the tech firms claims that using existing works to train AIs is fair use and does not need to be paid for.
https://www.youtube.com/watch?v=R5pFE85oAnA [8 minutes]
The tech bros are wallowing in almost unimagineable wealth: they can definitely afford to compensate copyright holders for using their work as training data. Alternatively, they can let copyright holders exclude their works from use in training data and compensate them for what they have used without permission.
I don't believe the tech companies have some kind of natural right to generate new works that are closely modelled on existing works without paying for their use of those works. The new works generated by humans are already pretty derivative in too many cases: we don't need AIs generating still more of the same.
There's a wealth of art (whether music, visual art, or literature) freely available in the public domain. Let them use that if they need large quantities of art to train their models.
--
Rhino