500 freaking billions

[SiliconWizard]

https://openai.com/index/announcing-the-stargate-project/

To get some perspective, that's almost twice the budget of the whole Apollo program, adjusted for inflation.

[XxMandragoraxX]

WoW! I hope that with this technology we will be able to conquer the Milky Way

[johnboxall]

I hope it doesn't gain self-awareness.

[XxMandragoraxX]

I don't think it is a system capable of living alone, it will always need the human. Basically, an AI is not going to be put to chop coal or control a nuclear power plant to obtain energy. It may decimate the human race, but there are decades left for that. Anyway humans are very adaptive and there are many of us, we will always win.

[Whales]

I hope it doesn't gain self-awareness.

Don't worry.

1. For many years (centuries?) authors have written stories about machines gaining self-awareness.
2. In recent years neural network research has increased and become a business.

The only link between these two things is marketing.  That's why they call these products "AI", they want you to think they're the same thing, despite having no other (non-marketing) links to each other.  Can you think of any other reason that these two are linked?

These algorithms are interpolation machines, really good at making answers to things within the bounds of their training data, but poor at extrapolating beyond that.  Their biggest practical applications are fun, pattern recognition and confusing legal systems (suddenly it's easy to do collusion, copyright infringement, working around examination rules, etc).

[SiliconWizard]

The algorithms are still nothing special - what makes them "powerful" is the gigantic amount of data, and that's why the new gold rush is data, data, and more data.

[tszaboo]

I hope it doesn't gain self-awareness.

There are much bigger and more urgent problems with AI and big data.
Companies already using it to maximize misery. How? By using it with dynamic pricing to find the maximum profit possible. And this is on our expense. For example hotels are using dynamic pricing, where they increase the room prices to extract as much money as possible from their customers. And they dynamically adjust the prices through some software, and all hotels are using this software. You cannot affor it anymore, but someone else will (while swearing a lot) and it's a win for the hotel. Why have 5 customers where you can have 4 with more profit?
And this is somehow not illegal, because you see, they don't sit together in one room, discussing how much the price should be, so it's not a cartel. They only do that with some API, so it's legal.
So that's what we using AI for.
Maximizing misery.

[golden_labels]

But they didn’t use ChatGPT to write that PR blah blah?

I’m not moved by $500 bn being put fed into the project. This isn’t much.

What is more worrying, is the other end. Investors want their money back and they want them with a huge profit. Guess who is going to pay for that.

[TimFox]

The funding, purpose, design, and control will be done by oligarchs.

[magic]

My tinfoil hat is whispering into my ear that "four years" and "in the United States" means "federal grants", considering recent events and The Man Who Must Not Be Named.

[TimFox]

This is perhaps closer to an intended large investment by Foxconn in a Wisconsin site, but this time the oligarchs are probably interested in doing it to meet their own goals.

[wilfred]

I hope it doesn't gain self-awareness.

I know you have tongue firmly in cheek.

I think of AI and more like artificial stupidity. It is a giant pattern, or word frequency database that relies on volume of data to provide probabilities of word relationships to create the appearance of intelligence. But it has no conceptual understanding and cannot do more than give the appearance of understanding.

Which is perhaps the nearest it actually gets to human intelligence.

It is a tool that will reach a limit of cost savings more than anything else pretty quickly and then we'll all sit back a lament what a colossal waste of money, time and effort it was as we tried to push it into too many aspects of life. It might be good for things like checking banking transactions for fraudulent looking activities checking internet searches for possible new disease outbreaks very early. But is that really intelligence?

But all trump cares about is doing it before the Chinese, which harkens back to beating the Russians in the 1960's. At least back then it felt like a real achievement.

[coppice]

"Give me $500B and I will create something. I've no idea what it might be able to do, but I'll create something.". Wonderful project proposal. 

[Gyro]

I hope it doesn't gain self-awareness.

If it does, it will probably die of embarrassment and performance anxiety.

[madires]

Is building power plants part of their project? Otherwise they won't be able to power their pseudo AI.

[golden_labels]

I think of AI and more like artificial stupidity. It is a giant pattern, or word frequency database that relies on volume of data to provide probabilities of word relationships to create the appearance of intelligence. But it has no conceptual understanding and cannot do more than give the appearance of understanding.

Which is perhaps the nearest it actually gets to human intelligence.

It is a tool that will reach a limit of cost savings more than anything else pretty quickly and then we'll all sit back a lament what a colossal waste of money, time and effort it was as we tried to push it into too many aspects of life. It might be good for things like checking banking transactions for fraudulent looking activities checking internet searches for possible new disease outbreaks very early. But is that really intelligence?

It isn’t going anywhere, because it already did establish its position even before the world heard about the few big generative networks.

Don’t confuse the entire branch of research and industry, with which the civilization is now intimately linked, and the two interwined happenings of recent years:

• A few specific products and their offshoots. All extremely marketed, which is why laypeople know about them.
• Making unfounded promises about such products’ imminent progress. About them turning into a concept taken from pop-culture, not the actual field.

With the ingress of opportunistists hoping for quick and easy money, it seems like yet another bubble. And I hope this bubble bursts. But similar to ocean trade, railroads, or internet — all three of which were the subjects of legendary bubbles — the technology itself appear to not be at risk of falling into obscurity.


The final question is rhetorical, isn’t it? But, together with the mention of getting near the human intelligence, it begs to be replied. And it can’t be answered for the same reason that underpins ignosticism. Everybody believes they know what intelligence is. But when used outside of a specific field, it almost exclusively refers to a nebulous notion with no tangible meaning. Itself rooted in now antiquated ideas about human psychology: the consequence being that even if one manages to outline some discernable shape, it remains nonsensical.

[iMo]

..$100billion investment now for the marketing, and for remaining $400billion they will order the AI "Prodigy" chips for the new datacenters.. 

[golden_labels]

The internet bubble bursted, because money were in large part spent only on marketing and making promises. And I don’t see OpenAI producing anything more than empty promises. Do they have some secret technology they never revealed? They better do, because I don’t find anything suitable for the task on their shelves. So far the company seems to me like a peddler saying they will turn a rug into a space rocket, if only I pour enough gold into the pocket.

Perhaps it’s a very elaborate and beautiful rug. We may even know legends of flying carpets, and of pharaohs falling out of rugs. But, pardon if I’m talking bullshit due to not being an expert in either rugs or rockets, in my perhaps naïve perception you can’t turn a rug into a rocket. In a similar fashion I don’t see how you turn a language model into “AGI,” whichever reasonable definition of that term we accept. I also argue I know more about algorithms and statistics than I do about either rugs or rockets.

The beauty of OpenAI’s promise may lie in their use use of term “AGI” and how susceptible this term is to goalpost shifting.

[wilfred]

The final question is rhetorical, isn’t it? But, together with the mention of getting near the human intelligence, it begs to be replied. And it can’t be answered for the same reason that underpins ignosticism. Everybody believes they know what intelligence is. But when used outside of a specific field, it almost exclusively refers to a nebulous notion with no tangible meaning. Itself rooted in now antiquated ideas about human psychology: the consequence being that even if one manages to outline some discernable shape, it remains nonsensical.

I wasn't posing a rhetorical question. Is AI really intelligence? Sure it is using very powerful (and power hungry) silicon to match input to a prior store of data it has structured, and will match new input datasets with in order to "make sense" of it for some particular function. And it may well do it usefully to quickly and cheaply mimic what a human may do more slowly, or may not easily do at all if time and volume of data is a both limited and overwhelming. AI may even provide some insight that a subset of humans may not even have thought of or expected. Again is that intelligence? A human presented with something unexpected might reasonably think yes it is. But is it? Whatever it was that an AI machine extracted from the data was always there. AI did not produce something that did not previously exist by virtue of the volume of human generated data that was used to "train" it. It is not like it created an original thought. Something that was NOT in the data used to train it.

 If, or when, AI produces output such as an idea that is novel (to the machine) then I think I might start to think of it as intelligent, but until that happens it is just a machine that regurgitates whatever has been input. And when it has become largely trained on other AI input data that itself was created from data originally based on human input data then there will be a crunch point in advancement in AI development.

If self driving cars are a example of a specific application of AI then I would expect AI having been trained on what an animal crossing the road looks like and having never been trained to recognise a kangaroo, which moves very differently to a cow or moose or dog, to also recognise one as an animal. Same goes for an overturned truck on the road. It may not have been trained by data showing many (or any) overturned trucks but I would still expect the car to not run into it.

It is when AI makes inferences from its training data to situations not in the data it was trained with. And furthermore it should recognise that and develop a hypothesis on how to learn more. Just as a human would do. In simple terms it should know that it does not know.

[coppercone2]

so you have linear computers, quantum computers and interpolation computers, similar to the UNN Von Braun

they are used to control the real AI ,but not the real AI 

[SiliconWizard]

The internet bubble bursted, because money were in large part spent only on marketing and making promises. And I don’t see OpenAI producing anything more than empty promises. Do they have some secret technology they never revealed? They better do, because I don’t find anything suitable for the task on their shelves. So far the company seems to me like a peddler saying they will turn a rug into a space rocket, if only I pour enough gold into the pocket.

Perhaps it’s a very elaborate and beautiful rug. We may even know legends of flying carpets, and of pharaohs falling out of rugs. But, pardon if I’m talking bullshit due to not being an expert in either rugs or rockets, in my perhaps naïve perception you can’t turn a rug into a rocket. In a similar fashion I don’t see how you turn a language model into “AGI,” whichever reasonable definition of that term we accept. I also argue I know more about algorithms and statistics than I do about either rugs or rockets.

The beauty of OpenAI’s promise may lie in their use use of term “AGI” and how susceptible this term is to goalpost shifting.

Yeah, well. Behind it there's SoftBank and Oracle. Isn't SoftBank very good at ruining companies usually? As to Oracle, don't get me started.

[TimFox]

In today’s news, oligarch Musk says the rival oligarchs in this consortium do not have sufficient resources.
Stay tuned.

[Analog Kid]

In today’s news, oligarch Musk says the rival oligarchs in this consortium do not have sufficient resources.
Stay tuned.

And yet Musk is the one who's warning us about the dangers of AI taking over the human race.
Does that show how schizoid the guy is?

[magic]

oligarch
oligarchs

I must say, one thing I like about this whole Orange Mango saga is that people are finally starting to see the system for what it is

And yet Musk is the one who's warning us about the dangers of AI taking over the human race.
Does that show how schizoid the guy is?

This was a thing for a long time between AI cultists like the OpenAI crowd. I don't even think Musk is the craziest of them. (I mean, he certainly is not, this title goes to Yudkowski et al, obviously).

[Marco]

Sam Altman proving himself the god emperor of grifting.

[magic]

You know, I could appreciate a skillful swindler for his "art", but these people are something else.

They are nutjobs who actually believe that they are here to revolutionize the world, that their technology will be the Next Big Thing, everyone will live happily ever after and they will get rich in the process, but that's OK because everyone will be rich. Not much different from Elizabeth Holmes or from Musk's Mars colonization program.

Silicon Valley is a cult, the whole place needs to be nuked from the orbit with no survivors.

[SiliconWizard]

They are nutjobs who actually believe that they are here to revolutionize the world, that their technology will be the Next Big Thing, everyone will live happily ever after and they will get rich in the process, but that's OK because everyone will be rich. Not much different from Elizabeth Holmes or from Musk's Mars colonization program.

Well, I see a small difference though: while most of these guys probably do believe that whatever they promote will benefit everyone, I don't think that's the case in particular for those in the "AI" field currently. Most of them have clearly stated, on the contrary, that AI was potentially very dangerous to humanity. So, no. They know it's not for the good of humanity. It's for something else.

Silicon Valley is a cult, the whole place needs to be nuked from the orbit with no survivors.

Given what's happening in California lately, you should probably refrain if you don't want to have the FBI at your doorstep. 

[magic]

But they are creating ethical AI. It's created by top experts in the field and it's explicitly trained not to misgender people, so it's safe. That's why it's so important that America, and not someone else, wins the race to AGI. And since you are still reading this, please donate to our Patreon, I really can't stress enough how important it is.

There will be peace, there will be UBI, there will be jobs for Americans, there will be unicorns. OpenAI AGI will make all of it possible, and then much more you haven't even dreamed of yet.

[SiliconWizard]

There will mostly be $500 billions injected into a few companies, so ultimately into a few pockets.
The unicorns are for the gullible who are ok with that much money (which doesn't grow on trees either, so ultimately much of it will come out of their pockets) being injected into this.

Even with the best intentions (which we know isn't the case), with $500 billions on the table, how ethical can you be? Heck, even the most detached buddhist monk would go bonkers with that much money on his monastery's table.

[magic]

Assuming that they will manage to raise those $500B.

But on second thought, maybe they are self-aware to some extent. I mean, this latest talk about "America first" is really hard to see as anything other than an opportunistic attempt at getting grants from Orange Mango administration. So maybe their talk about everybody slacking off on UBI and AI doing all work was the same, but for Obongo/Clinton era.

Either way, as I said, Vlad the Terrible and Winnie Pooh are the only people today with realistic means of solving the Silicon Valley problem

[XxMandragoraxX]

absolutely xD

[EEVblog]

There will mostly be $500 billions injected into a few companies, so ultimately into a few pockets.

[golden_labels]

I wasn't posing a rhetorical question. Is AI really intelligence?

If so, I believe I answered it. If that post was your reply to what I wrote, then it seems my entire point was missed. Either that or I fail to see how my words are reflected in your response.

I will repeat, just with more details, what I wrote above.

Word “intelligence” has no actual meaning, if used as a generic term. Just like to everybody, I myself not being excluded from the everybodies, to you it may seem you know what this “intelligence” is. But ask yourself: did you ever attach any precise properties to it? Properties that would either constrain it or be usable in discerning “intelligence” from other things? The same way you could do for a cat, for a house, for speech, for crying, for smell, even for such abstract and variable concepts as love, or things as ephemeral as “negation?” Did you? Can you?

There is an irony in the entire situation of talking about “intelligence” in machines, and in particular in generative LLMs. In majority of such debates and monologues, humans’ treatment of term “intelligence” is not different than what GPT networks do. Which, perhaps even more ironically, is a kind of the answer too.

Of course there are more strict uses of word “intelligence.” And if they are being referenced, the question is also much easier to answer. The problem is that it would likely not satisfy anybody.

Perhaps word “ignosticism” was the problem. I now see the Wikipedia article on it may be confusing to a reader, who didn’t know the concept earlier. There is a question about existence of god(s). The stance an ignosticist takes is: refusing to answer the question, until term “god” is properly defined. It isn’t an attempt to avoid dealing with it. Instead, it’s a reasonable reaction to how the statements about god(s) existence are made. It should also not be assumed to be an inherently atheistic view.

Well, I see a small difference though: while most of these guys probably do believe that whatever they promote will benefit everyone, I don't think that's the case in particular for those in the "AI" field currently. Most of them have clearly stated, on the contrary, that AI was potentially very dangerous to humanity. So, no. They know it's not for the good of humanity. It's for something else.

You missed one part of their statement. It’s dangerous to humanity. But they don’t say it has to be stopped, but that it has to be tightly controlled. And that “tight control” means preventing “the bad guys” from acquiring it. Which is achieved by keeping the technology in “trusted hands.”

Those are, what a surprise, their own hands. But that can’t be taken as doing so in bad faith. No, instead once again Hanlon’s razor does its job well. A common and simple human folly. The same thinking that makes dictators believe they’re doing good.

[SiliconWizard]

There will mostly be $500 billions injected into a few companies, so ultimately into a few pockets.

The club isn't that big. But yeah, I'm afraid I ain't in it.
Now I'm sure there will be crumbs of all that cash (so possibly still sizable crumbs) for many people working in the AI field in the coming years.
Not something I'm really interested in though. *Sigh*

[Marco]

There's already a ton of researchers, who are making way too much money to work particularly hard. There's already too much compute to begin with too, more compute doesn't help when company structure forces cowardice in how it's used. Yet more money is not going to propel the bloated AI companies to greatness, after Deepseek undoubtedly most private investors will wake up to that fact too. I'm going to agree with Elon, the private money simply isn't there.

[Analog Kid]

And yet Musk is the one who's warning us about the dangers of AI taking over the human race.
Does that show how schizoid the guy is?

This was a thing for a long time between AI cultists like the OpenAI crowd. I don't even think Musk is the craziest of them. (I mean, he certainly is not, this title goes to Yudkowski et al, obviously).

I guess I'm not calling Musk crazy so much as schizoid (not quite the same thing), since on the one hand he's not only promoting AI but depending on it for one of his ventures (Tesla), and yet on the other hand he's warning us about the potential Robot Uprising or whatever. Apparently he's decided that the benefits outweigh the risks--for now, at least.

I do think that the belief in the "singularity" that the AI fanbois imagine is truly meshuggah (crazy). They want to see the whole technology as omnipotent and omniscient. "All hail our AI masters!"

[Marco]

I do think that the belief in the "singularity" that the AI fanbois imagine is truly meshuggah (crazy). They want to see the whole technology as omnipotent and omniscient. "All hail our AI masters!"

That's more the trans/post-humanists, which are really not that mainstream. Sam and his fans somehow believe they can just let AI do all the work, while for the rest nothing much changes in society.

[magic]

Hardcore AI doomsdayers are scared of AI becoming "smarter" than humans and tricking humans relying on it for decision making into ultimately destroying humans for some nefarious purposes of the super AI.

This is obviously not a concern when "AI" is used for handwriting or speech recognition or even for driving cars, although the latter brings us to more real and immediate dangers such as half-assed AI vehicles causing road accidents.

A cynic could speculate that AI vendors tell those sci-fi stories to detract from the real damage caused by their products.

[Stray Electron]

But all trump cares about is doing it before the Chinese, which harkens back to beating the Russians in the 1960's. At least back then it felt like a real achievement.

  I agree. But the space race did cause some amazing technology to be developed.  I just hope that this boondoggle will also create some useful developments as well.

   Texas is already rather short of their power needs so locating this in Texas is going to REQUIRE that they create additional new and massive power sources in Texas.   My guess is that they will be nuclear.   Does anyone care to bet against that?

  FYI Trump claims that private investment is going to pay for all of this. Elon clearly thinks otherwise!

[Stray Electron]

You know, I could appreciate a skillful swindler for his "art", but these people are something else.

They are nutjobs who actually believe that they are here to revolutionize the world, that their technology will be the Next Big Thing, everyone will live happily ever after and they will get rich in the process, but that's OK because everyone will be rich. Not much different from Elizabeth Holmes or from Musk's Mars colonization program.

Silicon Valley is a cult, the whole place needs to be nuked from the orbit with no survivors.

   Not Silicon Valley, but Washington, DC!  SV is just a reflection of the nuttiness and irrational thinking and irresponsible behavior of the political elitists that rule the country.  And NOT just the current administration but most of them over the last 100 years or so.  They think that they're TinkerBell and that that can just wave their magic Congressional wands and everything will turn into their imagined Perfect world.   Does anyone remember LBJ's Great Society Program?  How's that coming? 

   For the ones of you that weren't even alive when that fisaco was started, here is a brief description from the internet.  Can anyone say that ANY these goals have been accomplished in the past 60 years, despite massive government spending? 

The Great Society was a series of domestic programs launched by President Lyndon B. Johnson in the 1960s aimed at eliminating poverty and racial injustice. Key initiatives included reforms in education, healthcare, and civil rights, significantly expanding the role of the federal government in social welfare.

[Analog Kid]

Yes, I well remember the Great Society, including such subprograms as Model Cities.
In fairness, it may have been an unachievable goal, but at least it was an honest one, sincerely undertaken to improve people's lives.
Can't say the same for "AI" at all.

And where's my backyard nuclear power plant?

[golden_labels]

I guess I'm not calling Musk crazy so much as schizoid (not quite the same thing), since on the one hand he's not only promoting AI but depending on it for one of his ventures (Tesla), and yet on the other hand he's warning us about the potential Robot Uprising or whatever. Apparently he's decided that the benefits outweigh the risks--for now, at least.

As I mentioned above, what they (he’s not alone) worry about is not the technology, but the technology being misused. And believe that this will never happen, as long as it stays completely under their control. A normal and widespread feature of Homo Sapiens. A mistake even easier to ignore, where strong financial, status, and power incentives are present.

Sam and his fans somehow believe they can just let AI do all the work, while for the rest nothing much changes in society.

And that is no different than any other major technological change, is it? Pasting an element in a picture one is used to, failing to notice the entire picture changes. The bane of all policymaking: adjusting one piece without realizing it triggers change in the entire intricate network of feedbacks and dependencies.

Hardcore AI doomsdayers are scared of AI becoming "smarter" than humans and tricking humans relying on it for decision making into ultimately destroying humans for some nefarious purposes of the super AI.

It matches the patterns set by fairytales we all hear as kids and how world is conveyed to us by adults. It has to be some malicious people, secret or not, the extraterrestials, the demons, or some uncontrollable technology. It’s critical that it is an entity, can have attached negative traits to it, and you don’t need much mental capabilities to understand the problem. Something a simple monkey brain may quickly swallow.

But there is a story even more frightening than any tangible monsters. Worse than any malevolent entity taking over control. It’s the story in which nobody and nothing has control. Guess which of the options is more likely. I got my first teaching on that matter over 20 years ago. Long before smortnets became a thing and indeed they’re nothing more than one more brick in this.

A cynic could speculate that AI vendors tell those sci-fi stories to detract from the real damage caused by their products.

Not a cynic, but an ignorant. The stories are much, much older. So is the associated fear. Vendors don’t have to do anything: it all arrived on a fertile ground, which was ready for harvest. No need to put effort in distraction either. People are doing this very well of their own accord.

(I assumed the common meaning of word “cynic,” not the proper philosophy)

[Stray Electron]

Yes, I well remember the Great Society, including such subprograms as Model Cities.
In fairness, it may have been an unachievable goal, but at least it was an honest one, sincerely undertaken to improve people's lives.
Can't say the same for "AI" at all.

And where's my backyard nuclear power plant?

   Coming soon to a backyard near yours. But don't worry it will be owned by some big energy consortium that will charge you a small fortune for your power needs. 

[Analog Kid]

Yes, I well remember the Great Society, including such subprograms as Model Cities.
In fairness, it may have been an unachievable goal, but at least it was an honest one, sincerely undertaken to improve people's lives.
Can't say the same for "AI" at all.

And where's my backyard nuclear power plant?

   Coming soon to a backyard near yours. But don't worry it will be owned by some big energy consortium that will charge you a small fortune for your power needs.

B-b-b-bbutt they told me it would be "too cheap to meter"?!?!?

[TimFox]

Yes, I well remember the Great Society, including such subprograms as Model Cities.
In fairness, it may have been an unachievable goal, but at least it was an honest one, sincerely undertaken to improve people's lives.
Can't say the same for "AI" at all.

And where's my backyard nuclear power plant?

   Coming soon to a backyard near yours. But don't worry it will be owned by some big energy consortium that will charge you a small fortune for your power needs.

B-b-b-bbutt they told me it would be "too cheap to meter"?!?!?

If I remember correctly, the original meaning of that phrase for commercial nuclear power systems was the assumption that the marginal cost of production for electrical power was negligible compared with the capital and maintenance costs of the reactors and distribution grid.  Therefore, the sensible pricing model would concentrate on peak demand (influencing the cost of the copper to connect you) rather than on actual consumption.
This contrasts with conventional power, where the cost of coal or gas to fuel the boilers correlated to the consumption of electrical power.
Historically, when Edison first went to capitalists to fund his new-fangled system, their first question was how to charge each consumer, whereupon he invented a suitable watt-hour meter.

[coppice]

If I remember correctly, the original meaning of that phrase for commercial nuclear power systems was the assumption that the marginal cost of production for electrical power was negligible compared with the capital and maintenance costs of the reactors and distribution grid.  Therefore, the sensible pricing model would concentrate on peak demand (influencing the cost of the copper to connect you) rather than on actual consumption.
This contrasts with conventional power, where the cost of coal or gas to fuel the boilers correlated to the consumption of electrical power.
Historically, when Edison first went to capitalists to fund his new-fangled system, their first question was how to charge each consumer, whereupon he invented a suitable watt-hour meter.

I think the quote was contracted from "it will be too cheap to meter,  but it will be $10k a month subscription to pay the capital costs." 

[TimFox]

See  https://www.nrc.gov/reading-rm/basic-ref/students/history-101/too-cheap-to-meter.html
Strauss was suggesting an analogy to city water systems, that often charged a fixed monthly fee rather than charge by the gallon.
(My house's account with the Chicago Water Department is fixed-rate:  I had seen too many stories about billing snafus associated with switching to actual water meters, which is now available as an option.)

[coppice]

See  https://www.nrc.gov/reading-rm/basic-ref/students/history-101/too-cheap-to-meter.html
Strauss was suggesting an analogy to city water systems, that often charged a fixed monthly fee rather than charge by the gallon.
(My house's account with the Chicago Water Department is fixed-rate:  I had seen too many stories about billing snafus associated with switching to actual water meters, which is now available as an option.)

Interestingly, the too cheap to meter approach for water has turned out to be a huge problem over much of the planet, and water meters are a growth industry.

[TimFox]

That's an important point:  I was objecting to misunderstanding of the original statement for a cheap laugh.
Our situation in Chicago, next to a huge body of fresh water, is quite different from other parts of the world.
Related questions include pricing of agricultural irrigation water versus industrial customers and domestic customers, as well as praciticability of desalinization or other technical solutions.

[coppice]

That's an important point:  I was objecting to misunderstanding of the original statement for a cheap laugh.
Our situation in Chicago, next to a huge body of fresh water, is quite different from other parts of the world.
Related questions include pricing of agricultural irrigation water versus industrial customers and domestic customers, as well as praciticability of desalinization or other technical solutions.

There are lots of interesting questions about charging for utilities, and what the bulk of people will find fair and reasonable. Researching this you get to ask interesting questions, which most people have never thought about. If you ask most people why they buy a book they answer something to the effect that "I wanted to read it". However, you don't need to buy a book to read it in most developed places. We have libraries that either have the book, or will get it for you if you wait a couple of weeks. Access to the contents of the book does not have to cost you anything. Buying a book is something more to do with convenient access to the contents, not buying the contents themselves. If you are at college its usually that demand for the book at the libraries available to you swamps the number of copies in the library system. If its a dictionary you may want it always at hand, for convenience. Questions like this run through most of the issues of pricing utilities, and can get quite complex in areas where there is strong competition encouraging people to experiment.

[TimFox]

Years ago, there were private lending libraries that charged reasonable fees to rent a book.
Big cities like Chicago have large public library systems, including a central library, regional libraries, and branch libraries.
Smaller US cities are not so blessed in the library department.
Some of my former colleagues objected to paying taxes for other people to read library books.

[coppice]

Years ago, there were private lending libraries that charged reasonable fees to rent a book.
Big cities like Chicago have large public library systems, including a central library, regional libraries, and branch libraries.
Smaller US cities are not so blessed in the library department.
Some of my former colleagues objected to paying taxes for other people to read library books.

Yeah, life is complex. Studying people's attraction to subscription vs pay per use models gets interesting. People generally like subscription models, as they like the predictability in their outgoings month by month. Beyond that you might expect high users to prefer subscription models, and low users to prefer pay per use. However, when you do market studies of a large number of people they have wildly inaccurate perceptions of whether they are in the high user or low user category.

[tszaboo]

See  https://www.nrc.gov/reading-rm/basic-ref/students/history-101/too-cheap-to-meter.html
Strauss was suggesting an analogy to city water systems, that often charged a fixed monthly fee rather than charge by the gallon.
(My house's account with the Chicago Water Department is fixed-rate:  I had seen too many stories about billing snafus associated with switching to actual water meters, which is now available as an option.)

Yes, and public transport is either too expensive and in a death spiral because less and less people use it.
Or the ticket sales barely cover the extra costs of the ticket system.

Going back to the original topic, the 500B is even a bigger joke now, that the Chinese released their super cheap AI models.
So the Deepseek was developed on apparently 2048 H100 GPUs, while Grok was made on 100.000. Who would have though that brute forcing everything is not the smartest way of doing things.

[Analog Kid]

Despite Tim Fox's putting a fine twist on the history of the phrase, I still contend that "too cheap to meter" as a cynical response to the optimistic "promise" of nuclear power is the appropriate one. That condition was never to come to pass, as the geniuses who saddled us with that technology never did their homework concerning the costs of the technology, including all the ancillary, hidden costs like those associated with the nuclear-fuel cycle, especially concerning things like waste disposal and plant decommissioning.

Which, BTW, is the real reason most places are hesitant to adopt more nuclear power generation: not so much the result of the agitation of people like me (I was an anti-nuclear activist in a former life), but simply the economic aspects of that technology. It just doesn't pencil out, despite Bill Gate's, et al, efforts to revive it with the "next generation" of reactors.

[SiliconWizard]

Going back to the original topic, the 500B is even a bigger joke now, that the Chinese released their super cheap AI models.
So the Deepseek was developed on apparently 2048 H100 GPUs, while Grok was made on 100.000. Who would have though that brute forcing everything is not the smartest way of doing things.

"Brute forcing" is pretty much the right definition for what we currently do with AI based on large models anyway.
Sure, there are ways to be more efficient, but that's still brute force.
It's a war on data.

[themadhippy]

that the Chinese released their super cheap AI models.

soon to be banned in the usa unless the owners flog it off cheap to a group of american investers

[Analog Kid]

that the Chinese released their super cheap AI models.

soon to be banned in the usa unless the owners flog it off cheap to a group of american investers

It's open source!
Good luck "banning" that.
My hat's off to the Chinese here.

[xrunner]

Nvidia shares sink as Chinese AI app spooks markets

US tech giant Nvidia lost over a sixth of its value after the surging popularity of a Chinese artificial intelligence (AI) app spooked investors in the US and Europe.

DeepSeek, a Chinese AI chatbot reportedly made at a fraction of the cost of its rivals, launched last week but has already become the most downloaded free app in the US.

AI chip giant Nvidia and other tech firms connected to AI, including Microsoft and Google, saw their values tumble on Monday in the wake of DeepSeek's sudden rise.

In a separate development, DeepSeek said on Monday it will temporarily limit registrations because of "large-scale malicious attacks" on its software.

https://www.bbc.com/news/articles/c0qw7z2v1pgo

[Marco]

soon to be banned in the usa unless the owners flog it off cheap to a group of american investers

Multiple US model as a service providers already provide it.

An unfortunate NVIDIA employee also asked for support for model support for their software on their public forum, because they wanted to use it internally for coding and they can't use external models. That post got removed and I assume they got a talking to.

[David Hess]

There are much bigger and more urgent problems with AI and big data.
Companies already using it to maximize misery. How? By using it with dynamic pricing to find the maximum profit possible. And this is on our expense. For example hotels are using dynamic pricing, where they increase the room prices to extract as much money as possible from their customers. And they dynamically adjust the prices through some software, and all hotels are using this software. You cannot affor it anymore, but someone else will (while swearing a lot) and it's a win for the hotel. Why have 5 customers where you can have 4 with more profit?
And this is somehow not illegal, because you see, they don't sit together in one room, discussing how much the price should be, so it's not a cartel. They only do that with some API, so it's legal.
So that's what we using AI for.
Maximizing misery.

If only there was a feedback loop which would attract capital investment in industries where profits are excessive, like building more housing and hotels.  We could call it the economics of a free market.

[David Hess]

Despite Tim Fox's putting a fine twist on the history of the phrase, I still contend that "too cheap to meter" as a cynical response to the optimistic "promise" of nuclear power is the appropriate one. That condition was never to come to pass, as the geniuses who saddled us with that technology never did their homework concerning the costs of the technology, including all the ancillary, hidden costs like those associated with the nuclear-fuel cycle, especially concerning things like waste disposal and plant decommissioning.

In the US, nuclear waste disposal was built into the pricing model from almost the beginning.  Nuclear power is taxed with the funds turned over to the government, which spent the money and left IOUs, like with Social Security and USPS pensions.

https://en.wikipedia.org/wiki/Nuclear_Waste_Policy_Act

Decommissioning is paid for in a similar way.

[tszaboo]

There are much bigger and more urgent problems with AI and big data.
Companies already using it to maximize misery. How? By using it with dynamic pricing to find the maximum profit possible. And this is on our expense. For example hotels are using dynamic pricing, where they increase the room prices to extract as much money as possible from their customers. And they dynamically adjust the prices through some software, and all hotels are using this software. You cannot affor it anymore, but someone else will (while swearing a lot) and it's a win for the hotel. Why have 5 customers where you can have 4 with more profit?
And this is somehow not illegal, because you see, they don't sit together in one room, discussing how much the price should be, so it's not a cartel. They only do that with some API, so it's legal.
So that's what we using AI for.
Maximizing misery.

If only there was a feedback loop which would attract capital investment in industries where profits are excessive, like building more housing and hotels.  We could call it the economics of a free market.

I think on some level more sinister things are happening. For sure someone is already working on AI models, where the feedback loop is people. What kind of information do we have to feed people to change their opinion about something. They already showed that something like a dozen likes are very good predictor on your personality. They could totally do a controlled experiment on just 1% of the user base, and see what sort of info will change your opinion on many things. Like how you vote, what brands you use, what sort of social issues you are interested in. How much you are willing to pay for things. Just do these games on 100.000 people, nobody will notice. Or if they do, "oh it's in beta" and "that's just hallucination".
There are more and more websites and online stores, that only show prices after login.
I really hope it's not all downhill from here.

[David Hess]

There are much bigger and more urgent problems with AI and big data.
Companies already using it to maximize misery. How? By using it with dynamic pricing to find the maximum profit possible. And this is on our expense. For example hotels are using dynamic pricing, where they increase the room prices to extract as much money as possible from their customers. And they dynamically adjust the prices through some software, and all hotels are using this software. You cannot affor it anymore, but someone else will (while swearing a lot) and it's a win for the hotel. Why have 5 customers where you can have 4 with more profit?
And this is somehow not illegal, because you see, they don't sit together in one room, discussing how much the price should be, so it's not a cartel. They only do that with some API, so it's legal.
So that's what we using AI for.
Maximizing misery.

If only there was a feedback loop which would attract capital investment in industries where profits are excessive, like building more housing and hotels.  We could call it the economics of a free market.

I think on some level more sinister things are happening. For sure someone is already working on AI models, where the feedback loop is people. What kind of information do we have to feed people to change their opinion about something. They already showed that something like a dozen likes are very good predictor on your personality. They could totally do a controlled experiment on just 1% of the user base, and see what sort of info will change your opinion on many things. Like how you vote, what brands you use, what sort of social issues you are interested in. How much you are willing to pay for things. Just do these games on 100.000 people, nobody will notice. Or if they do, "oh it's in beta" and "that's just hallucination".
There are more and more websites and online stores, that only show prices after login.
I really hope it's not all downhill from here.

There is a feedback loop like I described in economics, but the lack of a free market explains rising hotel rates, as well as rental and home ownership rates.  Nothing more sinister is required.

AI is being used to create a more efficient market revealing an already existing problem.  If you want lower rates, then increase the supply of hotel rooms, rentals, and housing.  This is a political problem having nothing to do with AI.  Of course the people (special interests) profiting from the current situation will have something to say about that, but the legislators are even more responsible for the situation.  Stop voting for them.

[Analog Kid]

In the US, nuclear waste disposal was built into the pricing model from almost the beginning..

Bullshit.
Provably, demonstrably false.
Up until about the 1960s the industry was still kicking the can of waste disposal down the road. There were not even any usable game plans on what to do with that shit right up until the Nuclear Waste Policy Act was passed in 1982.

I researched this, so I know what the hell I'm talking about here.

And don't even get me started on the Price-Anderson Act, which indemnifies the nuclear industry in the case of any catastrophic incident. Without that prop the industry never would have gotten off the ground in the first place.