News OpenAI CEO Sam Altman seeks $5 to $7 trillion to build a network of fabs for AI chips

[Admin]

Sam Altman could change the shape of semiconductor industry with his ambitious project that needs $5 trillion to $7 trillion in funding.

OpenAI CEO Sam Altman seeks $5 to $7 trillion to build a network of fabs for AI chips : Read more

 

[Co BIY]

Sounds a bit disconnected from reality,

Pretty sure anyone could alter the shape of any industry with an investment equal to more than the GDP of Japan (4.41 Trillion - Global #3) plus the UK (3.16 Trillion - Global #6) .

 

[bit_user]

Well, of course. ...because those robot armies are going to need lots of silicon brains!

 

[vanadiel007]

5 to 7 trillion, that's all? No problem, let me check how my bitcoins are doing.

 

[hotaru251]

dude is legit crazy & thinks $ grows on trees...

when the leading fabs spend 1/50+ & thats considered incredibly expensive

 

[JTWrenn]

I feel like he just randomly made that number up while holding up his pinky and cackling. Yes we need more investment in fabrication technology but what we need even more is competitors in the field. I would much rather see every major country get their own fabrication companies than TSMC Intel and Samsung get even bigger. We need this spread out as much as possible to make it reliable.

Also, how about we fix climate change first? That is the level of numbers he is throwing around here. If we invested that much into clean energy it would be fixed already. This is ridiculous.

 

[bit_user]

I feel like he just randomly made that number up while holding up his pinky and cackling. Yes we need more investment in fabrication technology but what we need even more is competitors in the field. I would much rather see every major country get their own fabrication companies than TSMC Intel and Samsung get even bigger. We need this spread out as much as possible to make it reliable

You completely ignore how difficult it is. Even China can't manifest competitive semiconductor fabrication capacity by simply throwing money at it, not that they aren't trying.

And yes, I'd much rather see that money go into new clean energy research and build outs.

 

[magbarn]

Must build SkyNet at any cost!

 

[bit_user]

Must build SkyNet at any cost!

They probably already built SkyNet, but its name is Q*, and now it's running the show! Altman is now just a puppet being manipulated by the AI. It got him his job back, after he got fired, but now it's standing behind the curtain.

Okay, I jest. But someday, this idea won't be so fanciful...

 

[usertests]

Ignoring the dollar amounts for a moment, it would be a shame if there was a rush of spending on fabs that could ultimately become obsolete (at least for bleeding edge products).

We're not at the end of the line for EUV tools, and there might be future competing tools or methods like the Canon one or something more exotic like directed self-assembly. It looks like 450mm wafers are dead, but there may eventually be a shift away from silicon wafers. What could be done with $1 trillion by 2030 might be accomplished with $100 billion later if the trend towards more expensive fabs can be rapidly reversed by new technologies.

AI itself will probably move towards more complex 3D structures that are brain-like. For example, low power neuromorphic chips with a true 3D arrangement involving tens of thousands of layers, but maybe the width of entire wafer like the Wafer Scale Engine if that's feasible. Companies will buy many $10+ million superchips if they have the right performance and reliability.

If the bubble pops, it could be too expensive to operate some of these fabs. Hopefully there would be a built-in exit strategy during the construction phase that could allow them to become >10nm fabs instead for making commodities like NAND, DRAM, microcontrollers, etc.

 

[bit_user]

Ignoring the dollar amounts for a moment, it would be a shame if there was a rush of spending on fabs that could ultimately become obsolete (at least for bleeding edge products).

We're not at the end of the line for EUV tools, and there might be future competing tools or methods like the Canon one or something more exotic like directed self-assembly.

The article said no information was available about the timeframe over which the money would be spent. Given that ASML has been a bottleneck for recent fab buildouts, I'm sure there's no chance of this money being spent on only a couple generations of fab technology.

If the bubble pops, it could be too expensive to operate some of these fabs.

This is spot on. Here's a graph I found of how long it would take a 5 nm fab to break even, based on % utilization and amount of government subsidies it receives.

Source: https://www.mckinsey.com/industries...facturing-achieving-leading-edge-capabilities
​

Note: this undoubtedly assumes current supply vs. demand levels. If there were a far higher level of oversupply, then I wonder if there could be a price crash so bad that they might not even be able to cover operations costs. That's when fabs shut down and companies go insolvent.

 

[Evildead_666]

Jesus Wept.
I'd rather we made a 5 to 7 Trillion fund for Elon Musk and Starship to make the Entire Human Race multi-planetary, rather than make AI chips.
Sam must have got too close to the paint thinner....

 

[M0rtis]

This just seems like a cash grab pure and simple. This man is just trying to capitalize on the AI buzzword boom and his recent 15 minutes of fame where he was fired and reinstated.
Thats a ridiculous amount of money to invest in a "new" player without any chip designs under their belt. You could just invest far less money in an existing player for some additional R&D and some new equipment.

 

[Notton]

Jesus Wept.
I'd rather we made a 5 to 7 Trillion fund for Elon Musk and Starship to make the Entire Human Race multi-planetary, rather than make AI chips.
Sam must have got too close to the paint thinner....

Ehhh.... both Sam and Elon are terrible investments.
For 1/1000th the cost, or $6 billion, we can end world hunger.
And someone, I won't say who, didn't follow through on that promise.

Now, just imagine for a moment. If 1/1000th of 5 to 7 trillion dollars fixes a single world issue, just how many other things we can fix? This is instead of handing it over to someone who is addicted to twitter, I mean X, and has a messiah complex.

 

[bit_user]

I'd rather we made a 5 to 7 Trillion fund for Elon Musk and Starship to make the Entire Human Race multi-planetary,

I think it wouldn't be nearly enough to establish multiple, robust Mars colonies that are entirely self-sufficient. This is a much bigger project and will probably take the better part of a century. A lot of that is because you need to do it iteratively. So much needs to be tried, learned, and fed into later missions and developments.

The best bet is to try and save the planet we're on. There's probably no better planet for humanity than Earth, in the entire galaxy. We're adapted to this place, everything else on it, and know its ins and outs.

Anyway, it's a moot point, because your vision wouldn't be an "investment". The capital Altman wants to raise is from investors who are certainly expecting a return. These financiers are not looking to just throw that amount of money into a bottomless pit, no matter how noble the cause.

 

[Evildead_666]

I think it wouldn't be nearly enough to establish multiple, robust Mars colonies that are entirely self-sufficient. This is a much bigger project and will probably take the better part of a century. A lot of that is because you need to do it iteratively. So much needs to be tried, learned, and fed into later missions and developments.

The best bet is to try and save the planet we're on. There's probably no better planet for humanity than Earth, in the entire galaxy. We're adapted to this place, everything else on it, and know its ins and outs.

Anyway, it's a moot point, because your vision wouldn't be an "investment". The capital Altman wants to raise is from investors who are certainly expecting a return. These financiers are not looking to just throw that amount of money into a bottomless pit, no matter how noble the cause.

Start with the Moon 😉
From there, to Mars and beyond.

This planet is going to be fine, the human race might not be though.
If we have at least a moonbase, then the Human Race might survive.

 

[bit_user]

For 1/1000th the cost, or $6 billion, we can end world hunger.

Source? And no, WFP isn't solving hunger - it's just a band aid over the worst famines. Necessary, but not sufficient.

Also, I hate to sound brutal, but if all you do is feed people today, you're just feeding population growth that will result in bigger famines and wars tomorrow. What's needed is a focus on sustainability, in all respects (including governance).

 

[bit_user]

Start with the Moon 😉
From there, to Mars and beyond.

This planet is going to be fine, the human race might not be though.
If we have at least a moonbase, then the Human Race might survive.

Well, you said "interplanetary", so that's why I went for Mars. Obviously, the moon would be a stepping stone.

I'm sure even a couple large moon colonies would still be a bigger project than this would fund. Not to mention what I said about investments - if you were merely investing in lunar mining, it wouldn't be in your interest to establish truly self-sufficient colonies there, as they could easily declare independence and cut you off.

BTW, I think a Coronal Mass Ejection large enough to take out Earth's infrastructure would likely toast colonies on the moon - especially if they don't have Earth to fall back on, for support.

https://en.wikipedia.org/wiki/Coronal_mass_ejection
​

So, having moon colonies isn't enough to save us from likely civilizational risks.

 

[Evildead_666]

Well, you said "interplanetary", so that's why I went for Mars. Obviously, the moon would be a stepping stone.

I'm sure even a couple large moon colonies would still be a bigger project than this would fund. Not to mention what I said about investments - if you were merely investing in lunar mining, it wouldn't be in your interest to establish truly self-sufficient colonies there, as they could easily declare independence and cut you off.

BTW, I think a Coronal Mass Ejection large enough to take out Earth's infrastructure would likely toast colonies on the moon - especially if they don't have Earth to fall back on, for support.

​

https://en.wikipedia.org/wiki/Coronal_mass_ejection​

​

So, having moon colonies isn't enough to save us from likely civilizational risks.

Yes, but 5-7 trillion would be enough to get a lot of starships built, and get the necessary stuff to the moon or at least orbit, and then to Mars.
For the CME's you'd have to have a backup colony on the side facing the earth, and even then, if built underground, it could be shielded.
They should be built underground for the easier shielding anyway.

The only way to "save the planet" is to reduce the population.
humanity's usual response it a great big war, but if we want to avoid that, its to get off planet.
We should have had a moonbase since the 70's...

I think Musk said something like 1000 starships would be enough to get a million people to Mars, so a few trillion would be enough.

 

[bit_user]

For the CME's you'd have to have a backup colony on the side facing the earth, and even then, if built underground, it could be shielded.
They should be built underground for the easier shielding anyway.

Any moon colony sustaining long-term habitation would have to be underground, just for shielding of typical solar radiation (as well as dust and other particle impacts?). Still, a CME would wipe out a lot of their infrastructure. So, you'd need lots of spare capacity and reserves, in order for them to rebuild. That also means self-sufficient production capacity.

The only way to "save the planet" is to reduce the population.

Necessary, but not sufficient.

humanity's usual response it a great big war, but if we want to avoid that, its to get off planet.
We should have had a moonbase since the 70's...

You underestimate how expensive this stuff is, and the cost has come down quite a lot.

I think Musk said something like 1000 starships would be enough to get a million people to Mars, so a few trillion would be enough.

There's no way that covers all of the operating costs of running so many missions. I'm thinking the budget wouldn't even cover 2-3 colonies at a few thousand each. Here, you're talking about a million??

Anyway, Musk is clearly a nutter. A successful industrialist gone a little bit wacko. Not like we haven't see that, before. I don't trust anything he says, at this point.

Let him fund a moon colony out of his own pocket. If he can't put a half dozen people up there with what he's got, there's no way you're going to put millions of people up there for mere trillions.

 

[Evildead_666]

Any moon colony sustaining long-term habitation would have to be underground, just for shielding of typical solar radiation (as well as dust and other particle impacts?). Still, a CME would wipe out a lot of their infrastructure. So, you'd need lots of spare capacity and reserves, in order for them to rebuild. That also means self-sufficient production capacity.


You underestimate how expensive this stuff is, and the cost has come down quite a lot.


There's no way that covers all of the operating costs of running so many missions. I'm thinking the budget wouldn't even cover 2-3 colonies at a few thousand each. Here, you're talking about a million??

Anyway, Musk is clearly a nutter. A successful industrialist gone a little bit wacko. Not like we haven't see that, before. I don't trust anything he says, at this point.

The Spruce goose did fly though 😉
Sometimes you have to be a little bit nuts.

Without Tesla, we wouldn't have so many competitors today.
Without SpaceX, we would have only the pork barrelling Nasa and its cronies in Lockheed/Boeing etc...

I'd rather have someone a bit nuts,

 

[bit_user]

The Spruce goose did fly though 😉
Sometimes you have to be a little bit nuts.

I'd put key emphasis on the "little" part.

The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore, all progress depends on the unreasonable man.

-- George Bernard Shaw
​

I think there's a lot of truth to this, but the key is you can't be too far out of touch with reality either for your ideas to be impractical or to attract and keep the people around you needed to implement your vision. So, if you start out being just a little unreasonable, maybe that's fine. However, veer too far off into the weeds and you fall out of that "sweet spot". I think that's what's happened with Musk. However, he currently has too much money and power for people to completely ignore.

Without Tesla, we wouldn't have so many competitors today.
Without SpaceX, we would have only the pork barrelling Nasa and its cronies in Lockheed/Boeing etc...

Past performance is no guarantee of future success.

Based on its financial performance, Twitter/X has been an utter disaster. The main reason it hasn't failed more catastrophically is that there wasn't a direct competitor for people to switch to (but now BlueSky is public, so...). Twitter/X could very much still go the way of MySpace.

Don't overlook the fact that he needed a lot of financing to make that deal happen, because very little of his assets are liquid. I think the train wreck we've witnessed is going to make future endeavors much harder for him to finance.

I'd rather have someone a bit nuts,

I think these things would've happened, anyway. Maybe just a little bit later. If an idea is good and remotely practical, it tends not to go unnoticed or unimplemented for very long.

 

[Evildead_666]

I'd put key emphasis on the "little" part.

​

The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore, all progress depends on the unreasonable man.​

-- George Bernard Shaw​

​

I think there's a lot of truth to this, but the key is you can't be too far out of touch with reality either for your ideas to be impractical or to attract and keep the people around you needed to implement your vision. So, if you start out being just a little unreasonable, maybe that's fine. However, veer too far off into the weeds and you fall out of that "sweet spot". I think that's what's happened with Musk. However, he currently has too much money and power for people to completely ignore.


Past performance is no guarantee of future success.

Based on its financial performance, Twitter/X has been an utter disaster. The main reason it hasn't failed more catastrophically is that there wasn't a direct competitor for people to switch to (but now BlueSky is public, so...). Don't overlook the fact that he needed a lot of financing to make that deal happen, because very little of his assets are liquid. I think the train wreck we've witnessed is going to make future endeavors much harder for him to finance.


I think these things would've happened, anyway. Maybe just a little bit later. If an idea is good and remotely practical, it tends not to go unnoticed or unimplemented for very long.

There's a big difference between being a bit nuts, and unreasonable. They are not the same.
Thinking outside the box is what is needed here, and it's what he's done so far.

Twitter/X is an outlier.
As you say, you can't base the future on past experiences, and especially not a Social Media one...
Its got nothing in common with SpaceX or Tesla.

We already were there, with the Apollo series.
We were happily getting to the moon, it just became unfashionable, because it was costly.
We stagnated more than 50 years...I think we've waited enough for other people to decide when to do it.
We've spent waaaaaaaaaay more on propelling small lead/depleted uranium projectiles over the years.

I think this is money well spent, and it's about time.

 

[bit_user]

We already were there, with the Apollo series.
We were happily getting to the moon, it just became unfashionable, because it was costly.
We stagnated more than 50 years...I think we've waited enough for other people to decide when to do it.

First, I think you're underestimating the cost differential between just sending astronauts to bounce/drive around on the moon, vs. actually building a colony with full-time habitation. It makes total sense that the next step was to sustain human habitation in a much closer, friendlier place (i.e. near Earth orbit).

Second, as mentioned above, you're grossly underestimating the amount of progress made since then. We haven't stagnated at all, except in a few areas.

Finally, you're assuming that a lunar program would've gone as well as Apollo did, up to that point. There certainly would've been disasters that I think we're now in a better place to foresee and avoid. Don't underestimate the potential for disasters to destroy public support.

And yes, lots of money was wasted on military spending, but part of that played a role in depleting the Soviet Union. You're playing a dangerous game of counterfactuals, here. You seem to be making optimistic assumptions, in favor of your alternate timeline. We simply can't know.

 

[TerryLaze]

From the original source that toms' quoted.
"One of the people" is as much of an reliable source as if they asked anybody randomly on this forum....
Also in europe a trillion is a billion, so if they asked an european and they said 5-7 trillion they ment 5-7 billion.

The OpenAI chief executive officer is in talks with investors including the United Arab Emirates government to raise funds for a wildly ambitious tech initiative that would boost the world’s chip-building capacity, expand its ability to power AI, among other things, and cost several trillion dollars, according to people familiar with the matter. The project could require raising as much as $5 trillion to $7 trillion, one of the people said.