Cory Doctorow's ultimate crooked point is that you're fucked if you don't embrace AI.
-
Cory Doctorow's ultimate crooked point is that you're fucked if you don't embrace AI. Local models, whatever. He compares technologies that have been proven valuable with a product that is being predicted to be valuable.
It's a variation of the assumption that everyone who hates AI hasn't *tried* it. If only they would give it a chance, they'd not be left behind.
-
Cory Doctorow's ultimate crooked point is that you're fucked if you don't embrace AI. Local models, whatever. He compares technologies that have been proven valuable with a product that is being predicted to be valuable.
It's a variation of the assumption that everyone who hates AI hasn't *tried* it. If only they would give it a chance, they'd not be left behind.
it's very deceptive to compare a sloppy sometimes-correct-for-massive-compute-cost machine to these things on the basis of "they all came from shitty people/places"
-
it's very deceptive to compare a sloppy sometimes-correct-for-massive-compute-cost machine to these things on the basis of "they all came from shitty people/places"
@fasterandworse what about refusing to run an LLM on my laptop because i find them intensely, mind-numbingly boring
-
Cory Doctorow's ultimate crooked point is that you're fucked if you don't embrace AI. Local models, whatever. He compares technologies that have been proven valuable with a product that is being predicted to be valuable.
It's a variation of the assumption that everyone who hates AI hasn't *tried* it. If only they would give it a chance, they'd not be left behind.
@fasterandworse just like we got left behind when we didn't buy into bitcoin and EFTs...
-
@fasterandworse what about refusing to run an LLM on my laptop because i find them intensely, mind-numbingly boring
@rose_alibi that's the thing. It's clear that a prominent personality type just does not find the things appealing. It's infuriating to see this pressure when the most basic rejection reason is "IDGAF" and that in itself is a career hindering position.
-
@rose_alibi that's the thing. It's clear that a prominent personality type just does not find the things appealing. It's infuriating to see this pressure when the most basic rejection reason is "IDGAF" and that in itself is a career hindering position.
@fasterandworse like... i made mini predictive/generative text models for my own amusement for years. and then it got boring so i stopped. i don't see why a bigger one is going to be somehow less boring.
-
@fasterandworse like... i made mini predictive/generative text models for my own amusement for years. and then it got boring so i stopped. i don't see why a bigger one is going to be somehow less boring.
@fasterandworse literally the centerpiece of my MFA show was a NLP model of the DSM. it's not a lack of experience it's a fundamental lack of interest in going backwards.
-
it's very deceptive to compare a sloppy sometimes-correct-for-massive-compute-cost machine to these things on the basis of "they all came from shitty people/places"
saying that AI is useful for things like coding might be true, but is it *valuable* for things like coding? That's a different story. It can take a long time for bad code to fuck everything up.
-
saying that AI is useful for things like coding might be true, but is it *valuable* for things like coding? That's a different story. It can take a long time for bad code to fuck everything up.
@fasterandworse this is the first thing I thought about when “vibe coding” started getting pushed in our organization. The legacy code this is already creating is enormous. Maybe the thinking is you can just rebuild stuff over and over again from scratch because it’s so fast to do? won’t you eventually completely lose your user base from multiple service interruptions (at a minimum) and catastrophic failures/harm to them (at the worst)?
-
R relay@relay.infosec.exchange shared this topic
