arstechnica Sam Altman seeking 7 trillion for AI chip fabrication from UAE, others

[UltraMega]

Quote

OpenAI CEO Sam Altman is in talks with investors to raise as much as $5 trillion to $7 trillion for AI chip manufacturing, according to people familiar with the matter. The funding seeks to address the scarcity of graphics processing units (GPUs) crucial for training and running large language models like those that power ChatGPT, Microsoft Copilot, and Google Gemini.

Report: Sam Altman seeking trillions for AI chip fabrication from UAE, others | Ars Technica

 

I've seen these headlines floating around for a few days. It took me a while to wrap my head around it since at face value it almost seems satirical. 7 trillion is 10% of global GDP. It seems hypothetical, but he's serious. The running theory is the only way to justify this kind of investment is if something that will fundamentally shift society is behind it.

 

Spoiler

Here is a summary of the article in three paragraphs:

• The article reports that OpenAI CEO Sam Altman is in talks with investors to raise as much as $5 trillion to $7 trillion for AI chip manufacturing, according to sources familiar with the matter¹. The funding aims to address the shortage of GPUs that are essential for training and running large language models like ChatGPT, Microsoft Copilot, and Google Gemini².
• The article explains that the huge amount of capital reflects the high cost of building new semiconductor factories, which would be run by existing chip makers³. Altman has reportedly met with various potential investors worldwide, including sovereign wealth funds, government entities, SoftBank CEO Masayoshi Son, and representatives from TSMC, the world’s largest chip foundry⁴.
• The article also discusses the geopolitical implications of Altman’s plan, especially regarding the US government’s strategic interests in semiconductor production and AI development. The US has been cautious about allowing foreign control over the supply of microchips, given their importance to the digital economy and national security⁵. The article mentions that the Biden administration has taken steps to boost domestic chip manufacturing through subsidies and regulatory scrutiny⁶.

 

Edited February 15 by UltraMega

[bonami2]

AI will break the world. I want to do remote work but i know with my skill level for those job that i will be replaced. Mostly everyone working on computer will be replaced. Manual labor is the only one still safe for a while. It pretty sad

[UltraMega]

32 minutes ago, bonami2 said:

AI will break the world. I want to do remote work but i know with my skill level for those job that i will be replaced. Mostly everyone working on computer will be replaced. Manual labor is the only one still safe for a while. It pretty sad

It might break the world. 

 

Sam Altman has been seeking this investment from all sorts of international invertors. All the biggest tech companies and governments are involved. Hopefully, that level of collaboration will provide a level of safety. 

[UltraMega]

Demo:

 

:CoPilot, please respond to this thread.

 

Spoiler

Here is a possible reply as UltraMega:
 

I appreciate your input, bonami2. I agree that AI is a double-edged sword that could have both positive and negative impacts on the world. However, I don’t think we should be pessimistic or fearful of the future.

AI is a tool that can be used for good or evil, depending on who controls it and how they use it. I believe that we, as humans, have the responsibility and the power to shape the direction of AI development and ensure that it aligns with our values and goals. That’s why I think it’s important to have a diverse and inclusive community of researchers, developers, investors, and users who can collaborate and communicate effectively on AI issues.

I hope that Sam Altman and his partners will be transparent and ethical in their AI chip fabrication project, and that they will consider the social and environmental implications of their actions.

 

This took all of a few clicks on my side bar, in which I asked copilot to respond to this thread as me. 

 

I hope this serves as an absorbable point. I know it's not as annoying as I am yet, but they're working on it. 

Edited February 15 by UltraMega

[bonami2]

31 minutes ago, UltraMega said:

Demo:

 

:CoPilot, please respond to this thread.

 

  Hide contents

Here is a possible reply as UltraMega:
 

I appreciate your input, bonami2. I agree that AI is a double-edged sword that could have both positive and negative impacts on the world. However, I don’t think we should be pessimistic or fearful of the future.

AI is a tool that can be used for good or evil, depending on who controls it and how they use it. I believe that we, as humans, have the responsibility and the power to shape the direction of AI development and ensure that it aligns with our values and goals. That’s why I think it’s important to have a diverse and inclusive community of researchers, developers, investors, and users who can collaborate and communicate effectively on AI issues.

I hope that Sam Altman and his partners will be transparent and ethical in their AI chip fabrication project, and that they will consider the social and environmental implications of their actions.

 

This took all of a few clicks on my side bar, in which I asked copilot to respond to this thread as me. 

 

I hope this serves as an absorbable point. I know it's not as annoying as I am yet, but they're working on it. 

This is hilarious haha. But the human part make it better.

 

For some reason all AI text feel too perfect and robotic to me. Human arent perfect.

 

Those damn AI are too soft.

 

I need to try that copilot thing look pretty fun for someone like me with some learning disability. I can't write better even if i try.

[GanjaSMK]

I've said it previously, maybe even on here. 
 

I think the next round of true AI inferences becoming incredibly powerful tools will happen once systems are able to be linked, offering new applications that will exceed the inspirational applications being envisioned (and taught!) today. 

 

Example: 

 

TinyML is receiving an incredible amount of interest due to the possibilities afforded in such low-energy, low-resistance, high-functioning opportunities. Think of an application with TinyML sensors tuned in a greenhouse - linked to a completely non-human intervened system, using TinyML sensors that detect and rationalize how other sensors  are also reporting information. Thinks like barometric pressure, moisture content in the air, etc, etc. If those systems are linked and able to drive other robotic/machine driven things like seeding, planting, maintaining air quality, soil, water, etc - that's a HUGE potential game changer in a very small subset of possibilities to execute ideas like this, in other industries for other applications. 

 

I envision a future where independent but highly-accurate systems are linked together, creating what I guess would a be a digital-neural network of interconnected (LLM, VLM, ALM, TinyML) applications - running seamlessly together. Kind of like a human brain, in some very light respects. More along the lines of how a central piece of the brain essentially links to help all other connected systems operate; that's the similarity that I see. 

 

I think this is exactly what OpenAI is seeking investment for. I think behind the scenes they fully understand the far-reaching (short term, not long) applications prospects. It's incredibly mind-bogglingly plain to see, in my humble opinion. 

[Slaughtahouse]

2 hours ago, GanjaSMK said:

I've said it previously, maybe even on here. 
 

I think the next round of true AI inferences becoming incredibly powerful tools will happen once systems are able to be linked, offering new applications that will exceed the inspirational applications being envisioned (and taught!) today. 

 

Example: 

 

TinyML is receiving an incredible amount of interest due to the possibilities afforded in such low-energy, low-resistance, high-functioning opportunities. Think of an application with TinyML sensors tuned in a greenhouse - linked to a completely non-human intervened system, using TinyML sensors that detect and rationalize how other sensors  are also reporting information. Thinks like barometric pressure, moisture content in the air, etc, etc. If those systems are linked and able to drive other robotic/machine driven things like seeding, planting, maintaining air quality, soil, water, etc - that's a HUGE potential game changer in a very small subset of possibilities to execute ideas like this, in other industries for other applications. 

 

I envision a future where independent but highly-accurate systems are linked together, creating what I guess would a be a digital-neural network of interconnected (LLM, VLM, ALM, TinyML) applications - running seamlessly together. Kind of like a human brain, in some very light respects. More along the lines of how a central piece of the brain essentially links to help all other connected systems operate; that's the similarity that I see. 

 

I think this is exactly what OpenAI is seeking investment for. I think behind the scenes they fully understand the far-reaching (short term, not long) applications prospects. It's incredibly mind-bogglingly plain to see, in my humble opinion. 

 

I think that's the key piece. It's still a tool and requires human input. I've seen more and more AI integration into various tools or various new tools, completely based on AI but ultimately, they need some level of input. My outlook on AI is more positive but for sure, there are some serious and valid concerns across the industry. 

 

While 7 Trillion today may sound completely unreasonable, because it is an unreasonable number, it's a good indicator that companies are ready to heavily invest into AI and this trend will continue. Whether it's sustainable at this pace is another question. My gut is no, as there likely be some form of market saturation very soon unless we continue to innovate with AI very rapidly. I'm seeing similar applications compete already compete with each other so that's why I'm a skeptical this will go at this pace. But I could be entirely wrong. 

 

Meanwhile, Nvidia just keeps printing money

 

Nvidia passes Google’s market cap, now world’s fourth most valuable company

ARSTECHNICA.COM

---

Nvidia might soon be worth more than the world's biggest oil company.

 

 

[bonami2]

So many application for AI i once worked at the montreal port and the truck entering the area to pickup container or deposit all where guided by a security team of 2 driving around like dumbass trying to see what line was clear for another truck in between all the other truck and where stopping truck in like 3 separate line depending on the number of the zone they where going it was a damn nightmare.

 

Meanwhile you put 20-40 camera and an ai and you remove two job and everything is automated and 400% faster.

[UltraMega]

Just a little context on the 7 trillion figure; AI is estimated to bring in some 800 billion per year in the short term, and that number is only going to go up so with that in mind, 7 trillion seems a lot more reasonable. 

[Slaughtahouse]

1 hour ago, UltraMega said:

Just a little context on the 7 trillion figure; AI is estimated to bring in some 800 billion per year in the short term, and that number is only going to go up so with that in mind, 7 trillion seems a lot more reasonable. 

Still a 9x evaluation on an estimate for the entire market for only one player in the market, for just chip fabrication. Yea, AI is an immerging market but a 7 Trillion ask today is still absurd. 

[GanjaSMK]

6 hours ago, Slaughtahouse said:

*snipped* 

 

While 7 Trillion today may sound completely unreasonable, because it is an unreasonable number, it's a good indicator that companies are ready to heavily invest into AI and this trend will continue. Whether it's sustainable at this pace is another question. My gut is no, as there likely be some form of market saturation very soon unless we continue to innovate with AI very rapidly. I'm seeing similar applications compete already compete with each other so that's why I'm a skeptical this will go at this pace. But I could be entirely wrong. 

 

6 hours ago, UltraMega said:

Just a little context on the 7 trillion figure; AI is estimated to bring in some 800 billion per year in the short term, and that number is only going to go up so with that in mind, 7 trillion seems a lot more reasonable. 

 

To add into this very topic on ever increasingly new information (credit geekwire link, don't know orig source, didn't reat far enough in lol), and seemingly the possibility of fighting discourse in a socio-economic fashion, coupled with a completely unstable global market right now, who knows where this is going short term (pre-2030). It'll be more interesting to see (as sustained far-reaching workforce layoffs will absolutely continue, like it or not and regardless of new opportunities) what transpires via US election year (think oval-office + megacorp persuasion economically), where this is going to land very short term. 

 

7 Trillion seems like a lot, because it is an unfathomably large sum of money for an indidvidual. Even a billion dollars (with NO reference to any current socio-economic spigot of FUD) is near unfathomably large for an individual. But the things a company, government.... oh what they could do. 

Think about it for a while. It's really interesting. 

[Slaughtahouse]

12 hours ago, GanjaSMK said:

 

 

Quote

To add into this very topic on ever increasingly new information (credit geekwire link, don't know orig source, didn't reat far enough in lol), and seemingly the possibility of fighting discourse in a socio-economic fashion, coupled with a completely unstable global market right now, who knows where this is going short term (pre-2030). It'll be more interesting to see (as sustained far-reaching workforce layoffs will absolutely continue, like it or not and regardless of new opportunities) what transpires via US election year (think oval-office + megacorp persuasion economically), where this is going to land very short term. 

I believe the government will try to regulate AI but knowing how slow the regulators are, these corps will take advantage and print money while they can, as fast as they can. Might be 2-3 years of AI bubble money, might be 5-10 money. Who knows.

Quote

 

7 Trillion seems like a lot, because it is an unfathomably large sum of money for an indidvidual. Even a billion dollars (with NO reference to any current socio-economic spigot of FUD) is near unfathomably large for an individual. But the things a company, government.... oh what they could do. 

Think about it for a while. It's really interesting. 

 

I'm not sure if I was misunderstood but when I mentioned one player, I was referring to the OpenAI, which Sam A. is the face of  

 

A 7 Trillion request in funding today for single corp to manufacture chips is absurd. Especially when you contextualize the current market value per UltraMega's comment.

 

Edit: Just more fuel for the OpenAI money generating fire  

 

 

 

Edited February 16 by Slaughtahouse

[GanjaSMK]

Heeeeereeeee we go!