Back

@GossiTheDog I disagree on the theft part, "copying is not theft" and all that.

I agree on hoping they burn to a pile of rubble (or better yet, get dismantled and their remains distributed and put to better use).

Please don't ask me to explain the tension between these two statements, i'm not sure i could right now :)

@GossiTheDog A huge problem in the US now is that total imbeciles are running the show, both government and tech. How did we get here?

@GossiTheDog
So, training an AI with copyrighted work is in a "legal gray area"? I've been thinking about setting a trap of sorts for AI scrapers in a website's Terms of Service/use. Something along the lines of "if an AI gets content from here, the company which owns said AI must pay, say, 100 million $ to the website owner."
Then make one post about some topic and insert a made-up word along with the made-up definition. Wait for bots to ingest the trap, and then go ask AI about that word. If it spits out the definition, you have proof that it breached TOS. I wonder if it is possible?

@GossiTheDog

Why don't commercial AI tools corporations, build in cost factor for buying access to critical data sets they need for training their models?

Ohh, I forget, they want free data in order to commercially exploit it and not share benefits.

This is better known as extraction economics(ie, capitalism)

@GossiTheDog indeed, and the worst part is, you can't easily find out the data used to train these models, so you can't easily collect evidence to be able to adequately sue them for infringement.
(Not to mention forced arbitration loopholes)

@GossiTheDog the whole copyright debate should end about as soon as people start copying Disney's animation 'styles'