jmoiron plays the blues
@jmoiron.bsky.social replied: People have commented (complimentary) that my writing does not sound like AI in the past, but in case anyone is wondering, given the topic, I didn't use AI to write or research this post. I write these for my own self improvement and self expression.
@jmoiron.bsky.social I desperately wanted to complete at least one of the many AI takes in my drafts, and this one was it.
I fear it's like a chocolate chip sourdough; it's got some rustic appeal and some tasty morsels, but does it really work? Let me know!
jmoiron.net/blog/bespoke...
jmoiron/giverny main 6cf20fb use an actual toggle control to toggle on/off push notifs, use icon to also hint at its state
@jmoiron.bsky.social replied: I'm not saying I think some kind of purely rational or technocratic anti-AI movement would be better. The LessWrong folks are fairly anti-AI and I think only Ezra Klein takes them seriously.
It's just tough. This is the stronger part of the argument, but it doesn't inspire quick...
@jmoiron.bsky.social replied: The corrosive effects of AI on society, learning, and the economy are more nuanced. They are terrifying, but they're a harder conversation to have and a harder battleground to win.
A lot of well meaning people think there's a way to have AI and address those concerns.
@jmoiron.bsky.social replied: Lots of people have basic experience asking ChatGPT a question or whatever, and they can't square that with the claims that these things are dangerous.
Cases of people being convinced by AI to do harm should have consequences, but they don't rise to the number of accidental pool...
@jmoiron.bsky.social replied: I am pretty anti-gun, and I am incredibly concerned about climate change.
Imagine if we were constantly sounding the alarms over the environmental impact of gun oil manufacturing, and when people looked into it, it had about the same impact as the production of bushings for gara...
@jmoiron.bsky.social replied: Personally, I think they're hostile because they want the extreme remedy to AI harm; for AI companies to be shut down and AI development and usage to become illegal, but without the clear, existential danger that is more obviously expressed in the weaker claims, this is a harder ...
@jmoiron.bsky.social Some anti-AI movement claims are strong and some are weak, but the weak ones are more prominent and imply more extreme reaction. Many people in that movement are hostile to folks pointing out flaws in those claims, even though persisting with them damages their credibility.
@jmoiron.bsky.social replied: I know people think existing law is clear and sufficient, but I don't think it is. Not everyone interpreting training as a fair use activity is doing so cynically.
I'd welcome new legal standards for copyright/training, it seems novel enough to warrant them.
@jmoiron.bsky.social replied: Scalzi's post was the quote-in-quote that spurred me to post about this yesterday.
I was originally going to comment that Robert and John's posts didn't seem in conflict to me, but Scalzi later made it clear his post meant what I thought it meant, so it was just a misinterpretat...
@jmoiron.bsky.social The specifics of this case don't appear to apply to training in general.
LLM AIs reproducing copyrighted works from their training set seems intrinsic, and it's well known the companies trained off of pirated collections.
But training in the abstract still seems like fair use t...