$5 to $7 Trillion? Developing cutting edge chip fabrication technologies is insanely expensive but that number still strikes me as far too high. Unfortunately, the article doesn't go into any detail on the allocation of funds for a capital raise that unprecedentedly massive. I don't doubt that Altman may have cited that range, my skepticism is more centered on the practical possibility of productively putting that much money to work in a relevant startup time frame (~five years). Maybe there's missing relevant detail, like perhaps he was referencing a cumulative total spend over a decade or longer.
Just for comparison TSMC's market cap is ~$500B, ASML (who makes the cutting edge fab machines) is ~$350B. NVidia itself is ~$1.7T but that also includes non-AI things. Even if you include buying smaller players in addition to those, you could theoretically outright own the entire relevant market for less than $3T. And that's at full de-risked, retail public market price in a very frothy time period. The new venture cost should be less, otherwise investors would be wiser to just buy the existing players with all their IP, proven know-how and market momentum.
Not to mention, you can’t just throw a large pile of money somewhere and create another ASML. It took them decades to get EUV working, as that tech is insanely sensitive and complex.
Unless you are outright buying some of their key people and their secrets. Or instead just licensing the tech from ASML. It would be a quick buck for ASML with little downside.
Apple and Microsoft are $3 trillion each, and they don't have fabs. If anyone has good reasons for getting into semiconductor manufacturing and has the pockets to do so, it's them, but they aren't. I'm pretty sure that it has something to do with risk, not money.
We're talking about chips: stuff optimized for inference and training. It may have a lot of memory, it may be able to perform a lot of operations, but it would be difficult for it to enforce high-level concepts like "woke ChatGPT".
Are you telling me that the investors who give sama $trillions don't expect influence on the final product?
Isn't this the definition of soft power? [0]
When NVIDIA sells GH200s to OpenAI, they expect nothing beyond the transaction. That's very different from getting trillions of dollars in investment from somebody to build chip fabs, etc.
Hysterical to me people on here in other threads would be knocking Altman for trying to raise huge money from people with huge money who have literally made it a goal to diversify into other areas.
> “Burn an enormous pile of (oil) money, to gain a large but smaller pile of (chip) money”
But continue having that pile of (chip) money after the oil and gas money runs out (either due to decarbonisation or due literally the resources running out).
I don't know why HN loves Sam Altman so much. Has he ever written a significant piece of software or invented anything? Has he ever built a significant, profitable business from the ground up? The guy was a fundraiser superstar when money was easy to get in the 2010s but I have yet to see technical or business competence sufficient to grant him the level of respect many in this community seem to grant him.
There are people who are technically competent or genuinely intelligent who can also raise funds.
The guy reminds me of Elizabeth Holmes or Eliezer Yudkowski of "lesswrong."
A sophomoric guy with minimal technical or business accomplishments who is just smart enough to tell rich people what they wanna hear to fuel his game.
There is an old term for this personality type. Snake oil salesman aka con artist.
Though Loopt was not a smashing success, it was moderately successful for a time.
Sam then advised start-ups and was president of YC.
He has been a key figure in OpenAI since it's inception, and based on the response from the employees when he was booted, I'd say they have a great respect for his leadership.
When Sama left YC for OpenAI, it was still 2.5-3 years from launching a product (I could be wrong on that). So he has lead the release of the fastest growing company.
So, I'd say yes, he has done some very significant things, and will likely continue to do them.
However, I also would be cautious of conflating founder with investor, which I see quite often. Sam is involved with a fusion company (I don't know which one), and I hear it spoken as if that is his company, but I believe he is just the investor. Maybe an active investor, but I agree, let's give credit to the people on the field, not those who buy the box seats.
No need. The point of a business is to make money not just burn it. It is that simple. It is kind of funny that you attempt to appeal to authority to some how invalidate this elementary concept. It doesn't. It really isn't impressive, from a business standpoint that a smooth talking huckster college dropout with no significant track record of significant business success can burn billions, now trillions of dollars of other people's money without having a viable business.
He's the new Elon Musk that people can idolize. I really don't understand the obsession with corporate venture capitalist vultures and trust fund babies. He's a glorified hype man. He curates a positive public image and makes dubious claims about the abilities of <insert new tech here> to drum up funding.
There is a large subset of tech oriented people who are extremely susceptible to sensationalism and treat these people like cult leaders to aspire to. Its just odd.
It's the way people are wired. They need a hero to look to. Now usually the outcome is better in the end when the hero is genuine not a mimic. I think "show, don't tell," is a good rule of thumb for judging character. It's even in the Bible as a parable of Jesus something like judge a vine by its fruit.
In the early 21st century it seems judgements of business acumen have become untethered from actually making money by building a real product that is in demand by the market. At a certain point the game of shuffling money around as a shell game will become unsustainable. That's already evidenced in the high inflation rate we have.
You can work backwards. For $5T investment, you need revenues of at least $500B. Nvidia revenues are currently ~18B/quarter or roughly $75B/yr. If you think 5X growth in AI in next 5 year is possible but current supply chain cannot deliver then all these totally makes sense.
To me, AI is starting to replace convectional search within early adopters already. The new search is not firing query to get references but rather converse to get answer. In my vicinity, 90% of the population has not even caught up on this progress, let alone change of muscle memory which often takes half decade.
And search is still just one of many segments. So, market has 10X growth potential from here with fair certainty. Although, as it often happens, we will overshoot expectations that causes crash followed by more regularized growth.
Fun thing is that a ton of wild predictions that created Internet bubble in 2000 actually were rather timid. Internet did changed everything and delivered way beyond expected during the bubble. It was only a matter of time.
Sam Altman is just repeating the early days Elon playbook.
He tries to keep the media coverage saturated with bullshit which avoids scrutiny on the wrong things and has the convenient side effect of burning attention otherwise drawn towards things like Google’s Gemini Product launch.
The fact that things are basically verbatim reported without any critical thinking shows he’s right, it works. Trump employed the same strategy too.
It takes advantage of our terminally broken media ecosystem (no incentive to call bullshit), personality cult and attention economy.
The only way to read Silicon Valley is to look at actions, not words.
This seems like a smart move. OpenAI needs to run models more cheaply for profitability. He wants to start a hardware venture in parallel with the goal of selling better hardware to OpenAI. If it fails, it's not his money that was lost. If it succeeds, he has two successful ventures on his hands.
Anybody know what happened to George Hotz' hardware company? Or did he give up on it after 2 weeks like his Twitter frontend job?
Like I used to do to my parents? Whenever I wanted 5 trillion I'd always ask for 7 trillion so they can feel good about teaching me a lesson about the value of money, and they would feel 2 trillion richer and I'd still be able to build my nuclear aircraft carriers.
Just for comparison TSMC's market cap is ~$500B, ASML (who makes the cutting edge fab machines) is ~$350B. NVidia itself is ~$1.7T but that also includes non-AI things. Even if you include buying smaller players in addition to those, you could theoretically outright own the entire relevant market for less than $3T. And that's at full de-risked, retail public market price in a very frothy time period. The new venture cost should be less, otherwise investors would be wiser to just buy the existing players with all their IP, proven know-how and market momentum.