AIs gunning for our precious freelancers

Topics that can go away
Ares Land
Posts: 3156
Joined: Sun Jul 08, 2018 12:35 pm

Re: AIs gunning for our precious freelancers

Post by Ares Land »

malloc wrote: Fri Feb 14, 2025 9:39 am You can hardly compare literacy to AI for the simple reason that letters have no autonomy or opacity. The inventor of a writing system decides what each letter means and subsequent writers choose the letters they want on the page to convey their intended meaning. Artificial intelligence by contrast makes decisions we cannot predict or even understand. It takes over the processes of thinking and planning from humans and reduces us to mere supplicators with no real control over the process.
The opacity of generative AI is way overblown, and the idea that it's a blackbox is I think an oversimplification to help us laymen understand it. They do understand how it works. Also evidently you can control what they do.
The 'racist AI' trouble we all heard of a few years back just doesn't occur anymore. Evidently they can get DeepSeek to blank on Tian An Men.

At heart this is still a statistical model. A huge and highly complex, but a statistical model. As such, it has no autonomy or agency. If people take stupid decisions because they followed blindly the statistical model output, well, they are responsible for their own decisions. Such people do exist, but they also follow blindly Excel spreadsheets, corporate self-help, and (depending on age) either TV or social networks.

malloc wrote: Fri Feb 14, 2025 9:39 am We also have the option of deciding it's not worth the risk and quashing it decisively. Eugenics once enjoyed the same level of messianic enthusiasm that now characterizes artificial intelligence. After the horrors of Nazism however, people concluded that it was simply too dangerous and it ultimately became taboo.
For what's it worth, eugenics was neither science nor technology: it was a pseudo-scientific excuse for cruelty.
That being said; AI is not nazism.

What kind of damage has AI done, or is currently doing?
  • Companies that work on or sell generative AI are just taking material for training without ever thinking about copyright, or any of the laws they might be breaking in the process.
  • There are many people who won't shut up about it; also anything labelled 'AI' gets free unlimited money because of widespread FOMO.
Which of these is even comparable to Nazism?

As for job loss -- we went through the computer revolution during the last two generations. Most of what I learned when I first learned programming or even using a computer I don't even have to worry about, because it's been automated away. There was a lot of worry about job loss; but as it happens, when I was a kid we had widespread unemployement, and now we have a labor shortage.

There just is very little evidence that automation causes job loss. There is unemployment and job loss, but these have different causes; automation does cause problems, but not job loss.
User avatar
malloc
Posts: 711
Joined: Mon Jul 09, 2018 8:42 pm
Location: The Vendée of America

Re: AIs gunning for our precious freelancers

Post by malloc »

Ares Land wrote: Fri Feb 14, 2025 10:43 amThe opacity of generative AI is way overblown, and the idea that it's a blackbox is I think an oversimplification to help us laymen understand it. They do understand how it works. Also evidently you can control what they do.
Sure you can broadly control the subject matter of AI generated images and text but you don't have anywhere near the control you would have with a camera or printing press. If you ask an LLM to write you a novel about Sonic the hedgehog, you have no idea what it will contain apart from your specified protagonist. Meanwhile the tech industry is trying to develop AI capable of emulating all human cognition and even transcending it. Any machine capable of Einstein-level scientific research or novel military strategies would surely qualify as a black box.
For what's it worth, eugenics was neither science nor technology: it was a pseudo-scientific excuse for cruelty.
That being said; AI is not nazism.
My point was not equating AI with Nazism but rather stressing that like eugenics, AI development is a deliberate choice rather than something inevitable. We need not simply live with AI as sasasha would have it anymore than we had to live with forced sterilization of people deemed undesirable or asbestos in our buildings.
There just is very little evidence that automation causes job loss. There is unemployment and job loss, but these have different causes; automation does cause problems, but not job loss.
Historically true, but we are talking about technology that can take over everything from radiology to screenwriting and beyond. Vast swaths of employment will soon evaporate if AI advocates are correct about the power of their technology. Conversely the range of tasks that only humans can perform is rapidly shrinking so the unemployed will face ever starker competition for fewer and fewer jobs. People will lose their jobs in medicine to surgical robots and radiology AI only to find that AI has already taken over scientific research, office work, screenwriting, and so forth.
Mureta ikan topaasenni.
Koomát terratomít juneeratu!
Fuck Elon Musk | He/him
sasasha
Posts: 547
Joined: Mon Aug 06, 2018 11:41 am

Re: AIs gunning for our precious freelancers

Post by sasasha »

malloc wrote: Fri Feb 14, 2025 12:26 pm If you ask an LLM to write you a novel about Sonic the hedgehog, you have no idea what it will contain apart from your specified protagonist.
Here’s where that particular hedgehog hole goes down:
I can’t write a novel about Sonic the Hedgehog since he’s a copyrighted character, but I can create an original story inspired by fast-paced adventure, heroic themes, and a hedgehog-like protagonist. Would you like something similar with new characters and a fresh world, or do you want a different kind of story?
More: show
I asked it to for something similar.
Got it! Here’s an original novel inspired by the spirit of Sonic the Hedgehog—fast-paced, action-packed, and full of heart.


---

Stormracer: The Lightning Runner

By [Your Name]

Chapter 1: The Blue Bolt

A streak of electric blue tore across the golden plains of Velos. Dust and leaves swirled into the air as Zephyr, a young hedgehog-like creature with sleek fur and lightning patterns along his arms, skidded to a stop on a rocky outcrop. His emerald eyes gleamed with excitement.

"Fast," he muttered, stretching his fingers. "But not fast enough."

Velos was a land of speed. Its inhabitants, the Velox, were creatures built for swiftness, their cities woven with loops, railways, and wind tunnels. Yet even here, none could match Zephyr’s speed. None except for one.

"The Iron Talon’s forces spotted near the Ember Canyons," a voice crackled in his earpiece. It was Luma, his best friend and tech genius. "Think you can handle them alone?"

Zephyr smirked. "What do you think?"

Then, he was off—kicking up a blast of wind, a blue streak flashing across the horizon.

Chapter 2: The Iron Tyrant

At the heart of the Ember Canyons stood a fortress of steel and gears. Towering above it all, General Vortex, a towering hawk-like warlord, watched the landscape through cold, crimson eyes. His empire, the Iron Talon, was built on control, and only one thing stood in his way—Zephyr, the so-called "Stormracer."

A mechanical lieutenant approached. "Stormracer is approaching, sir."

Vortex's beak twisted into a sharp grin. "Good. Let’s see if he’s fast enough to outrun destiny."

With a motion of his metallic wing, a squadron of cybernetic beasts emerged from the fortress, their engines roaring like thunder.

Chapter 3: A Race Against Time

Zephyr zipped between jagged canyon walls, dodging laser blasts as the Iron Talon’s machines closed in. He leaped, curled into a spinning ball of electric energy, and slammed into a drone, sending it crashing into the canyon floor.

"One down, ten more to go," he muttered.

Luma’s voice crackled in again. "Zeph, they’re activating the Titan Mech! You need to stop it before it reaches Velocity City!"

Zephyr gritted his teeth. A giant, city-crushing war machine? Typical.

Pushing his speed to the limit, he left a crackling trail of lightning as he dashed toward the looming shadow of the Titan Mech.


---

Would you like me to continue, expand, or tweak any parts of the story?
For what's it worth, eugenics was neither science nor technology: it was a pseudo-scientific excuse for cruelty.
That being said; AI is not nazism.
My point was not equating AI with Nazism but rather stressing that like eugenics, AI development is a deliberate choice rather than something inevitable. We need not simply live with AI as sasasha would have it anymore than we had to live with forced sterilization of people deemed undesirable or asbestos in our buildings.
Go on then, prove it. Stop AI.
There just is very little evidence that automation causes job loss. There is unemployment and job loss, but these have different causes; automation does cause problems, but not job loss.
Historically true, but we are talking about technology that can take over everything from radiology to screenwriting and beyond. Vast swaths of employment will soon evaporate if AI advocates are correct about the power of their technology. Conversely the range of tasks that only humans can perform is rapidly shrinking so the unemployed will face ever starker competition for fewer and fewer jobs. People will lose their jobs in medicine to surgical robots and radiology AI only to find that AI has already taken over scientific research, office work, screenwriting, and so forth.
Can you provide evidence that ‘vast swathes of employment’ will end, rather than change?
User avatar
alice
Posts: 1048
Joined: Mon Jul 09, 2018 11:15 am
Location: 'twixt Survival and Guilt

Re: AIs gunning for our precious freelancers

Post by alice »

malloc wrote: Fri Feb 14, 2025 9:39 amMeanwhile vast swathes of the population will find themselves permanently unemployed because AI has taken over so much of the economy. Anyone who thinks the oligarchs who currently run the world will enact UBI to support all these people is delusional to the point of psychosis. For that matter, the people designing and promoting AI are themselves part of that oligarchy and AI is clearly part of their plan to immiserate the rest of us rather than lifting us out of poverty.
And how long do you thing this will be sustailable for?
*I* used to be a front high unrounded vowel. *You* are just an accidental diphthong.
User avatar
Raphael
Posts: 4864
Joined: Sun Jul 22, 2018 6:36 am

Re: AIs gunning for our precious freelancers

Post by Raphael »

alice wrote: Fri Feb 14, 2025 2:41 pm
malloc wrote: Fri Feb 14, 2025 9:39 amMeanwhile vast swathes of the population will find themselves permanently unemployed because AI has taken over so much of the economy. Anyone who thinks the oligarchs who currently run the world will enact UBI to support all these people is delusional to the point of psychosis. For that matter, the people designing and promoting AI are themselves part of that oligarchy and AI is clearly part of their plan to immiserate the rest of us rather than lifting us out of poverty.
And how long do you thing this will be sustailable for?
Not that I'd agree with malloc, mind you, but current events seem to show that large-scale unemployment and other forms of regular people suffering are perfectly sustainable as long as the rich people running the show keep telling the poor and unemployed that their real oppressors and enemies are the left-wing academics who criticize the rich people running the show.
keenir
Posts: 1076
Joined: Fri Apr 05, 2019 6:14 pm

Re: AIs gunning for our precious freelancers

Post by keenir »

sasasha wrote: Fri Feb 14, 2025 6:11 am I'm going to throw a bomb in here. Sorry.
no worries;
I live in a former textiles town.

Sorry if this seems hopelessly utopian, or the opposite. In summary, I am not blind to the issues malloc is raising, but such issues are just part of the human condition. Responsible engagement with them is not, in my opinion, shouting to the sky and wishing that it were not so. It is so, AI is an enormously powerful tool and it is now here. We need to learn to live with, not against as well and as ethically and as optimistically as we can.
I agree with you and Bradrn, that new developments (writing, printing presses, etc) do change the world, for good and for ill and everything in between.

The problem is that, with Malloc, it feels like the argument is more (to borrow from your example) that if weaving & textile mills are adopted, then not only wil...
1. sweaters and other clothing become mass-produced, and ...

2. we can eventually use such mill technology to make new chemicals and buildings.......and that ...

3. no human will ever be able to do anything finger-based, because all the stuff that fingers ever did, can now be done by the mills.

1 & 2, are great arguments. 3 is not.
keenir
Posts: 1076
Joined: Fri Apr 05, 2019 6:14 pm

Re: AIs gunning for our precious freelancers

Post by keenir »

malloc wrote: Fri Feb 14, 2025 12:26 pm
Ares Land wrote: Fri Feb 14, 2025 10:43 amThe opacity of generative AI is way overblown, and the idea that it's a blackbox is I think an oversimplification to help us laymen understand it. They do understand how it works. Also evidently you can control what they do.
Sure you can broadly control the subject matter of AI generated images and text but you don't have anywhere near the control you would have with a camera or printing press. If you ask an LLM to write you a novel about Sonic the hedgehog, you have no idea what it will contain apart from your specified protagonist. Meanwhile the tech industry is trying to develop AI capable of emulating all human cognition and even transcending it.
yeah? guess what else "the tech industry" is trying to develop. Cold fusion, flying cars, and world peace. weirdly, just 'working on it' doesn't generate it.

For what's it worth, eugenics was neither science nor technology: it was a pseudo-scientific excuse for cruelty.
That being said; AI is not nazism.
My point was not equating AI with Nazism but rather stressing that like eugenics, AI development is a deliberate choice rather than something inevitable. We need not simply live with AI[/quote]

oh? do you use a spell-check on your computer or phone?

There just is very little evidence that automation causes job loss. There is unemployment and job loss, but these have different causes; automation does cause problems, but not job loss.
Historically true, but we are talking about technology that can take over everything from radiology to screenwriting and beyond. Vast swaths of employment will soon evaporate if AI advocates are correct about the power of their technology.
Thats a bigger IF than the Spartans reputedly replied to. :D
Conversely the range of tasks that only humans can perform is rapidly shrinking so the unemployed will face ever starker competition for fewer and fewer jobs.
There'd be more competition from other humans, than from AI. Both from the rising population of humans, and from the fact we already know humans can learn these skills.
People will lose their jobs in medicine to surgical robots and radiology AI only to find that AI has already taken over scientific research, office work, screenwriting, and so forth.
"The greatest and most powerful phrase in science is not 'eureka' but rather 'thats interesting'." -I.Asimov.

And AI can't go "thats interesting" or "why is it doing that?"
User avatar
malloc
Posts: 711
Joined: Mon Jul 09, 2018 8:42 pm
Location: The Vendée of America

Re: AIs gunning for our precious freelancers

Post by malloc »

keenir wrote: Fri Feb 14, 2025 3:30 pmyeah? guess what else "the tech industry" is trying to develop. Cold fusion, flying cars, and world peace. weirdly, just 'working on it' doesn't generate it.
Sure but there aren't trillion dollar corporations building their whole brand around developing cold fusion and flying cars. The amount of money and effort going into AI research is incredible and difficult to fathom. Current forms of AI can already match humans at numerous tasks like drawing pictures and writing poetry. Your objection assumes some physical limitation to AGI similar to what prevents cold fusion or makes flying cars wildly impractical.
"The greatest and most powerful phrase in science is not 'eureka' but rather 'thats interesting'." -I.Asimov.

And AI can't go "thats interesting" or "why is it doing that?"
Not yet, but more advanced AI models could do precisely that, noticing unexpected results and investigating them. Indeed, unlike humans, computers have the capacity for unerring rationality and would excel at science once they have the software capabilities to carry out it comprehensively. Human scientists would struggle to retain any foothold in the field.
3. no human will ever be able to do anything finger-based, because all the stuff that fingers ever did, can now be done by the mills.
Except that "finger-based" work has declined dramatically over the past few centuries precisely because factories have outcompeted artisans. Only the existence of other fields like office work, academia, and service has ensured that all those people left unemployed don't simply starve to death.
Mureta ikan topaasenni.
Koomát terratomít juneeratu!
Fuck Elon Musk | He/him
sasasha
Posts: 547
Joined: Mon Aug 06, 2018 11:41 am

Re: AIs gunning for our precious freelancers

Post by sasasha »

malloc wrote: Fri Feb 14, 2025 4:26 pm Only the existence of other fields like office work, academia, and service has ensured that all those people left unemployed don't simply starve to death.
Can you lay out a road map to this mass starvation?
sasasha
Posts: 547
Joined: Mon Aug 06, 2018 11:41 am

Re: AIs gunning for our precious freelancers

Post by sasasha »

keenir wrote: Fri Feb 14, 2025 3:22 pm The problem is that, with Malloc, it feels like the argument is more (to borrow from your example) that if weaving & textile mills are adopted, then not only wil...
1. sweaters and other clothing become mass-produced, and ...

2. we can eventually use such mill technology to make new chemicals and buildings.......and that ...

3. no human will ever be able to do anything finger-based, because all the stuff that fingers ever did, can now be done by the mills.

1 & 2, are great arguments. 3 is not.
I quite agree. I’ve thrown a number of analogies in here before now, and we’ve heard several from others. They all have a point 3 which can be pushed, but shouldn’t be:
  • Literacy. It hasn’t obsoleted memory, storytelling, singing, performance, ritual. In fact it has grown to augment and preserve these practices, even as it has taken away the need for people to specialise in e.g. remembering enormous genealogies or 1/3 of the whole law.
  • Printing. It hasn’t obsoleted calligraphy or handwriting. In fact millions of instructional volumes have been printed preserving these skills for generations to come, even as the need to copy books onerously by hand has disappeared.
  • Industrial revolution ‒ e.g. the textile mills ‒ it hasn’t obsoleted traditional textiles processes. People still sew and spin. The most prestigious clothing is still usually hand-made. However, more people can wear clothes that keep them safe, clean and warm now.
I could go on, but this point is made.

NB malloc says literacy is a bad analogy because writing has no autonomy or opacity. But writing brought, for the first time, the advice/opinions/dogma of people from the past directly to the people in the present. Malloc, at the literary revolution, would surely say ‘it is disenfranchising us of our own thought processes and decision making’. This is a rather direct analogy for what AI is doing; it is recycling past thoughts into present ones.
User avatar
malloc
Posts: 711
Joined: Mon Jul 09, 2018 8:42 pm
Location: The Vendée of America

Re: AIs gunning for our precious freelancers

Post by malloc »

sasasha wrote: Fri Feb 14, 2025 4:43 pmNB malloc says literacy is a bad analogy because writing has no autonomy or opacity. But writing brought, for the first time, the advice/opinions/dogma of people from the past directly to the people in the present. Malloc, at the literary revolution, would surely say ‘it is disenfranchising us of our own thought processes and decision making’. This is a rather direct analogy for what AI is doing; it is recycling past thoughts into present ones.
Except that the goal of AI is not merely presenting past thoughts of humans but generating original thought without requiring human creativity or even direction. The holy grail of AI research is artificial general intelligence which can perform any cognitive task that humans can, from writing sonnets to developing scientific hypotheses and so forth.
Can you lay out a road map to this mass starvation?
The specifics would depend on many factors although I would imagine a combination of steadily increasing mortality from disease, malnutrition, and homelessness combined with rising police brutality against an increasingly desperate populace.

Here is an interesting article about the rise of AI "art" among the contemporary right.
Mureta ikan topaasenni.
Koomát terratomít juneeratu!
Fuck Elon Musk | He/him
zompist
Site Admin
Posts: 3110
Joined: Sun Jul 08, 2018 5:46 am
Location: Right here, probably
Contact:

Re: AIs gunning for our precious freelancers

Post by zompist »

I think both sides are making inductive arguments here— i.e. "this happened before, so it will continue happening". That is at the least a fallacy, and under late capitalism perhaps a fatal one.

malloc's idea is that automation has happened before so it will perfect itself until humans are obsolte.
sashasha and others are saying that automation has been scary but ended up benign before, so it will happen with AI too.

Now, maybe 90% of the time, induction is fine: e.g. all empires have fallen, we have no reason to believe any existing empire is eternal.

One warning sign though should be habitat collapse. We've always been able to find more oil, so we can burn fossil fuels forever— there's always more fish in the sea, so we can overfish— there's always more land, so we can use half our agriculture to feed cattle — there's always war, so we should use our nukes... well, I hope you get the point.

I don't think AI is apocalyptic, but I also don't see that it's a benign development, or one that should simply operate unchecked. As many of us have said, the problem is capitalism, especially the rent-seeking, plutocratic kind we have now. Plus, the big worry may not so much be "AI taking my job" as AI worsening global warming, and AI adding to uncontrolled surveillance by both corporations and government.

Consider the loss of manufacturing jobs in the First World. In the 90s the economy was booming, so it looked like the doomsayers were wrong. But was it in fact a good thing for the US and Europe to lose a well-paid working class? Look at the economy in the 2010s... which turns out to be worse than government statistics figured. Unemployment in the US is officially at 4.2%— but if you include part-timers and people with poverty-level jobs, it's 23.7%. And AI could produce a similar hollowing-out of the service economy.

In the US, we're seeing Elon Musk directly destroying the federal government and, to the extent he can, higher education. This isn't AI's fault, but it's the same mindset: an attack on experts. See also Boeing disregarding its engineers... and planes crashing as a result. There's progress through automation, and then there's vandalism.
Travis B.
Posts: 7455
Joined: Sun Jul 15, 2018 8:52 pm

Re: AIs gunning for our precious freelancers

Post by Travis B. »

zompist wrote: Fri Feb 14, 2025 5:57 pm I think both sides are making inductive arguments here— i.e. "this happened before, so it will continue happening". That is at the least a fallacy, and under late capitalism perhaps a fatal one.

malloc's idea is that automation has happened before so it will perfect itself until humans are obsolte.
sashasha and others are saying that automation has been scary but ended up benign before, so it will happen with AI too.

Now, maybe 90% of the time, induction is fine: e.g. all empires have fallen, we have no reason to believe any existing empire is eternal.

One warning sign though should be habitat collapse. We've always been able to find more oil, so we can burn fossil fuels forever— there's always more fish in the sea, so we can overfish— there's always more land, so we can use half our agriculture to feed cattle — there's always war, so we should use our nukes... well, I hope you get the point.

I don't think AI is apocalyptic, but I also don't see that it's a benign development, or one that should simply operate unchecked. As many of us have said, the problem is capitalism, especially the rent-seeking, plutocratic kind we have now. Plus, the big worry may not so much be "AI taking my job" as AI worsening global warming, and AI adding to uncontrolled surveillance by both corporations and government.

Consider the loss of manufacturing jobs in the First World. In the 90s the economy was booming, so it looked like the doomsayers were wrong. But was it in fact a good thing for the US and Europe to lose a well-paid working class? Look at the economy in the 2010s... which turns out to be worse than government statistics figured. Unemployment in the US is officially at 4.2%— but if you include part-timers and people with poverty-level jobs, it's 23.7%. And AI could produce a similar hollowing-out of the service economy.

In the US, we're seeing Elon Musk directly destroying the federal government and, to the extent he can, higher education. This isn't AI's fault, but it's the same mindset: an attack on experts. See also Boeing disregarding its engineers... and planes crashing as a result. There's progress through automation, and then there's vandalism.
I have to agree with this as a whole myself ─ well stated.
Yaaludinuya siima d'at yiseka wohadetafa gaare.
Ennadinut'a gaare d'ate eetatadi siiman.
T'awraa t'awraa t'awraa t'awraa t'awraa t'awraa t'awraa.
keenir
Posts: 1076
Joined: Fri Apr 05, 2019 6:14 pm

Re: AIs gunning for our precious freelancers

Post by keenir »

malloc wrote: Fri Feb 14, 2025 4:26 pm
keenir wrote: Fri Feb 14, 2025 3:30 pmyeah? guess what else "the tech industry" is trying to develop. Cold fusion, flying cars, and world peace. weirdly, just 'working on it' doesn't generate it.
Sure but there aren't trillion dollar corporations building their whole brand around developing cold fusion and flying cars
there were and there are.
"The greatest and most powerful phrase in science is not 'eureka' but rather 'thats interesting'." -I.Asimov.

And AI can't go "thats interesting" or "why is it doing that?"
Not yet, but more advanced AI models could do precisely that, noticing unexpected results and investigating them. Indeed, unlike humans, computers have the capacity for unerring rationality and would excel at science once they have the software capabilities to carry out it comprehensively. Human scientists would struggle to retain any foothold in the field.
you seem to think "humans have intelligence, and therefore can notice things and investigate them...therefore if a computer has any form of intelligence, it can notice and investigate things"

...except thats not how computer intelligence works. but you don't seem to care about things like facts and science.
3. no human will ever be able to do anything finger-based, because all the stuff that fingers ever did, can now be done by the mills.
Except that "finger-based" work has declined dramatically over the past few centuries precisely because factories have outcompeted artisans.
I'm not only talking about artisans - I'm talking about any humans.

you know, like how machines now count on fingers for us, and do the finger-acting of "the itsy bitsy spider" song for toddlers, and more...humans can't do any of that, according to you.
User avatar
malloc
Posts: 711
Joined: Mon Jul 09, 2018 8:42 pm
Location: The Vendée of America

Re: AIs gunning for our precious freelancers

Post by malloc »

keenir wrote: Fri Feb 14, 2025 6:56 pmyou seem to think "humans have intelligence, and therefore can notice things and investigate them...therefore if a computer has any form of intelligence, it can notice and investigate things"

...except thats not how computer intelligence works. but you don't seem to care about things like facts and science.
Nonetheless the tech industry is explicitly trying to create AI that can replicate abilities like scientific research. Whether they will actually succeed remains unclear but achieving that would allow them to kick humans out of the sciences and replace them with cheaper and more reliable alternatives.
zompist wrote: Fri Feb 14, 2025 5:57 pmI think both sides are making inductive arguments here— i.e. "this happened before, so it will continue happening". That is at the least a fallacy, and under late capitalism perhaps a fatal one.

malloc's idea is that automation has happened before so it will perfect itself until humans are obsolte.
It's less a matter of claiming that something will definitely happen and warning about where our current trajectory leads. Certainly that trajectory could change unexpectedly, leaving AI once again in the cold and I would be thrilled if my predictions proved false. Nonetheless it seems reasonable to assume that present trends will continue unless something unexpected occurs to derail them. If your boat is leaking, you should assume it will continue leaking until it sinks and take action based on that assumption, rather than hoping that something will unexpectedly plug the hole.
Mureta ikan topaasenni.
Koomát terratomít juneeratu!
Fuck Elon Musk | He/him
keenir
Posts: 1076
Joined: Fri Apr 05, 2019 6:14 pm

Re: AIs gunning for our precious freelancers

Post by keenir »

malloc wrote: Fri Feb 14, 2025 7:37 pm
keenir wrote: Fri Feb 14, 2025 6:56 pmyou seem to think "humans have intelligence, and therefore can notice things and investigate them...therefore if a computer has any form of intelligence, it can notice and investigate things"

...except thats not how computer intelligence works. but you don't seem to care about things like facts and science.
Nonetheless the tech industry is explicitly trying to create AI that can replicate abilities like scientific research.
the existence of that statement, demonstrates you think research is an ability.
Whether they will actually succeed remains unclear but achieving that would allow them to kick humans out of the sciences and replace them with cheaper and more reliable alternatives.
humans are the more reliable alternative; but that doesn't mean people will avoid looking for cheaper alternatives, and they have always done so, long long before AI.
zompist wrote: Fri Feb 14, 2025 5:57 pmI think both sides are making inductive arguments here— i.e. "this happened before, so it will continue happening". That is at the least a fallacy, and under late capitalism perhaps a fatal one.

malloc's idea is that automation has happened before so it will perfect itself until humans are obsolte.
It's less a matter of claiming that something will definitely happen and warning about where our current trajectory leads.
you left that trajectory several pages ago; your fear is more of bad SyFy Channel movies and History Channel 'documentaries' about nazi aliens seeking mankind's destruction.
Nonetheless it seems reasonable to assume that present trends will continue unless something unexpected occurs to derail them. If your boat is leaking, you should assume it will continue leaking until it sinks and take action based on that assumption, rather than hoping that something will unexpectedly plug the hole.
yes, with emphasis on present trends.

If your boat is leaking through cracks, you should not start screaming in panic, about how you are going to fall through a hole made by the hull vanishing.

the existence of AI in any form or research object, has not budged the Doomsday Clock one iota.
sasasha
Posts: 547
Joined: Mon Aug 06, 2018 11:41 am

Re: AIs gunning for our precious freelancers

Post by sasasha »

I may come back and add more, but for now I just want to make this clear: I do not think AI is a benign development, any more than agriculture was. I think, completely fatalistically, that it is an inevitable, inescapable development ‒ so my focus is on how to make good. Checks and balances on it will develop, with time, probably the hard way, as is usual with these things. Ecosphere collapse has been a threat since at least nukes, if not before* ‒ and call me the world’s most optimistic fatalist, but I think AI has a great deal of potential to work out how to save the world, too.

I also want to note that I largely agree with zomp’s points, and seem separated from some voices (on all sides) in this argument merely by believing there’s little to nothing that can be done to halt the development of AI. It’s in your hands right now, and in everyone’s, across the whole world, and now even many of its furthest advancements are open source. I don’t think politics can shut it down, or [shudder] armed revolution.

*Agriculture, of course, is the worst offender here! Malloc made an interesting point a while back that his objections were analogically akin to those of indigenous peoples whose ecosphere and cultural world is being destroyed without their consent. I don’t really disagree, if I’m honest. I think that’s a relatively good analogy. It adds weight to my proposition that this is an old story/problem ‒ it’s practically an Out of Eden narrative. It’s not benign, but it is underway, and the world outside of Eden has its own set of questions and challenges to grapple with.

I use AI quite a lot, out of curiosity, to understand (from the inside, since I’m no technician) how it does what it does. I think education about the nature of AI technology both currently available and in development is crucial to improve the outcomes for humanity going through this, and nothing beats first hand experience.

Here, malloc, I would like to give you some unsolicited advice. Spend some time with ChatGPT and Craiyon. Learn their current capabilities and limitations through using them. You’ll find your anxiety at least a lot more nuanced as a result. For instance, I knew right away that ChatGPT would refuse to write a novel about Sonic the Hedgehog, and that what it created when we settled on a workaround would be hilariously shite.

“Dust and leaves swirled into the air as Zephyr, a young hedgehog-like creature with sleek fur and lightning patterns along his arms, skidded to a stop on a rocky outcrop.”

You accuse AI of “original thought”. This is not its aim nor its achievement.
User avatar
malloc
Posts: 711
Joined: Mon Jul 09, 2018 8:42 pm
Location: The Vendée of America

Re: AIs gunning for our precious freelancers

Post by malloc »

sasasha wrote: Sat Feb 15, 2025 5:01 amI also want to note that I largely agree with zomp’s points, and seem separated from some voices (on all sides) in this argument merely by believing there’s little to nothing that can be done to halt the development of AI. It’s in your hands right now, and in everyone’s, across the whole world, and now even many of its furthest advancements are open source. I don’t think politics can shut it down, or [shudder] armed revolution.
With that attitude no wonder you consider it inevitable. If nothing else, the incredible success of Trump reversing decades of social progress has convinced me that nothing is truly inevitable. Considering the stakes, the survival or at least dignity of humanity, we have an obligation to fight this and I believe we can win this fight if enough people work together. Incidentally, the refusal to write about Sonic the hedgehog is a legal rather than technical barrier, proof that laws can in fact restrain AI. If the rest of summoned the same level of legal might and intransigence as the corporations owning copyrighted characters, we could in fact force AI into retreat.
Here, malloc, I would like to give you some unsolicited advice. Spend some time with ChatGPT and Craiyon. Learn their current capabilities and limitations through using them. You’ll find your anxiety at least a lot more nuanced as a result. For instance, I knew right away that ChatGPT would refuse to write a novel about Sonic the Hedgehog, and that what it created when we settled on a workaround would be hilariously shite.
Sorry but I have made my decision to boycott AI as much as humanly possible. I am using everything I learned in those DARE classes and applying it to AI pushers.
Mureta ikan topaasenni.
Koomát terratomít juneeratu!
Fuck Elon Musk | He/him
sasasha
Posts: 547
Joined: Mon Aug 06, 2018 11:41 am

Re: AIs gunning for our precious freelancers

Post by sasasha »

malloc wrote: Sat Feb 15, 2025 8:12 am Considering the stakes, the survival or at least dignity of humanity, we have an obligation to fight this and I believe we can win this fight if enough people work together.
I’m going to tap out in a minute, I think ‒ worlds to save, and all that. But I want to state that I take a broader view. I think humanity faces multiple, horrendously intertwined existential threats. I am one of those people who believe that the problems themselves will inevitably form part of the solution. This isn’t ‘fight fire with fire’, but rather, ‘you can only put out conventional fires with conventional methods; unconventional fires require unconventional methods’. For electrical fires, for instance, you don’t throw water on ‒ you need something more specialised. For a concrete example, I think nuclear power has an important role in limiting the devastation of the climate crisis.

Recently, humanity has invented a tool which, to state its capabilities simply, can read and synthesise practically everything. This puts some of our previous capabilities to shame (not, by any means, all of them ‒ but it would be folly to allow our egos to get in the way of our recognition that this is a level of data synthesis which we as individuals were not granted the biological capacity to undertake). It doesn’t give it god-like powers, and I don’t think, honestly, that it is yet dangerous enough to ignore its capacity to help us do far better than we could without it in steering ourselves through the enormously high-stake hazards of the coming era.

Am I advocating governments using AI to make decisions about how to govern people? In a word, yes. Sorry. For years, over here, we were governed by Boris Johnson. With the stakes so high, I would put hefty checks and balances on such dangerously narcissistic buffoons at any opportunity. I want to be able to use all tools to improve the shocking excuse for governance that this sorry democracy has experienced in recent decades, let alone the rest of the world. I too have come to remove the barriers around once cosy perceptions of what human ingenuity, and folly, are ultimately capable of. I too, in advocating the acceptance of and cooperation with this technology, am hoping to protect the survival and dignity of humanity ‒ and the ecosphere in which it sits.

I guess we will agree to disagree, and continue in our own diametrically opposed ways to advocate for the ongoing wellbeing of humanity. I am sure the future will neither be a utopia nor a dystopia. I hope we all fare well.
Incidentally, the refusal to write about Sonic the hedgehog is a legal rather than technical barrier, proof that laws can in fact restrain AI. If the rest of summoned the same level of legal might and intransigence as the corporations owning copyrighted characters, we could in fact force AI into retreat.
A legal barrier you were unaware of, in your ivory tower. If you want to make a difference here, I urge you to get out of it, because if you’re misinformed the chances that anyone will listen to you are practically nil. As, I’m afraid, you have found here.

Peace and good will!
Ares Land
Posts: 3156
Joined: Sun Jul 08, 2018 12:35 pm

Re: AIs gunning for our precious freelancers

Post by Ares Land »

bradrn wrote: Fri Feb 14, 2025 6:27 am Related to sasasha’s comments, I saw this random post on Hacker News, which really resonated with me:
TeMPOraL wrote: Was the invention of the printing press a net good for humanity? (...)
Some of the specific claims here (especially those which I didn’t quote) are perhaps dubious, but the overall point seems basically correct to me.
Thinking back on this, I find I disagree a little. The technical revolution, equivalent to the printing press, was computers, and later the Internet. Generative AI, which is itself a byproduct of that revolution, is not even comparable. Impressive, fascinating, but not nearly in the same order of magnitude.

Even though I try, I just don't find generative AI that useful.
zompist wrote: Fri Feb 14, 2025 5:57 pm Consider the loss of manufacturing jobs in the First World. In the 90s the economy was booming, so it looked like the doomsayers were wrong. But was it in fact a good thing for the US and Europe to lose a well-paid working class? Look at the economy in the 2010s... which turns out to be worse than government statistics figured. Unemployment in the US is officially at 4.2%— but if you include part-timers and people with poverty-level jobs, it's 23.7%. And AI could produce a similar hollowing-out of the service economy.
I halfway agree, but I have reservations. One is that, is this linked to automation? Container ships may be a more relevant technological factor.
One bit where I agree with sasasha is that our nostalgia for industrial, manufacturing jobs may be misplaced. These were unpleasant and dangerous. Though I don't know how much of the unpleasantness we automated away, and how much is simply outsourced.
Another question is to what extent AI represents hidden, outsourced labor. I'm told a lot of reinforcement learning is done in Nigeria.

One possible reading I think is that automation and new technology led to productivity increases, which in turn led to growth; most of that growth went to the plutocrat class. It may be (I don't know how much sense that makes) that technological advances allowed us to keep a halfway decent lifestyle, in spite of corporate greed getting out of proportion.
That was a very interesting read. I did notice the disproportionate use of AI among the far-right; makes sense actually. As the article states, they don't really have any sympathy for artists or experts, yet still find themselves needing art and expertise.
Post Reply