CEO @TheFuturumGroup @FuturumEquities | Co-Founder @TheSixFiveMedia & @Signal_65 | @Forbes @MarketWatch Contributor | 7x Best-Seller | Dad x3 | Disclosures👇
Austin, TX
Joined December 2010
- Tweets14K
- Following3.7K
- Followers78.7K
- Likes35.2K
Great dinner in Raleigh tonight with @Lenovo CEO @YuanqingYang. The TLDR is AI has not been canceled. Demand is 📈
As they say… “Everyone is talking their book.”
I like Elon and of course I respect Elon, he’s one of the best entrepreneurs on the planet however nothing he says about neoclouds should be taken seriously.
SpaceX is essentially the largest neocloud (with hyperscaler aspirations)… so naturally he wants/needs other neoclouds to struggle with financing so he can swoop in and steal their demand/customers and/or he wants to buy their land/power thus fast tracking his capacity expansion plans at $SPCX otherwise it’s going to be very hard for him to hit 20GW in the next 2 years.
Ramp data says spend on AI flattening out but let’s not conflate that for usage which is surging. The spend is more likely inflecting than stalling as exponentially more tokens are being used to meet increasing demand on better performing models.
Execution will tell the story in the neocloud and LPS space. 👀
@elonmusk may be right that we'll see unfinished data centers. But not all developers operate the same way.
$WULF follows a disciplined approach:
1. Identify sites with demonstrated customer interest.
2. Secure power from investment grade utilities.
3. Lock down specs, construction costs and schedules through GMP EPC contracts.
4. Secure long-term customer commitments backed by investment grade credit.
5. Arrange financing before committing to major construction.
We may move dirt early, but we don't build on spec.
I suspect we'll see projects stall because of inadequate capital, weak counterparties, execution misses, or regulatory hurdles. For experienced infrastructure developers with strong balance sheets, that creates opportunity.
What others call stranded assets, we may call a development pipeline. @PSInvestor @GavinSBaker
The fungibility of Compute and Power is being under discussed right now. No compute or available power is going to go to waste between now and the end of the decade. 💡
I think a lot of folks are interpreting this as the end of neoclouds as a whole.
More realistically there will be a small subset of winners divided between true neoclouds and land power shell (LPS) developers that will thrive and grow.
And then there will be consolidation of smaller outfits that will bring the numbers down from hundreds two a dozen or two.
On the neocloud side there is room for $CRWV $NBIS $IREN and maybe 1-2 more to prosper.
On the LPS side there are many. I remain bullish on $CIFR and $WULF here. This segment will all but surely consolidate. 👀
Daniel Newman retweeted
Yes, the path is quite clear.
Lower inference cost -> Exponential usage -> infrastructure growth -> increased worker productivity -> economic (GDP) growth. 💡
CEO of Amazon, Andy Jassy, on the cost of inference:
"The cost of inference has come down quite a bit in the last few years, but it is still an order of magnitude too expensive, in my opinion, for it to be used as pervasively as possible. If you are asking enterprises to rewrite their applications that use CPU today with AI accelerators, if the difference in those costs are very significant. The amount of incremental benefit they have to get from their AI accelerators on there is a much higher bar to want to rewrite those apps."
"If you can take the cost of inference down, that bar is much lower and all applications are going to be better and smarter and more capable with AI accelerators and inference. That is a really important part for customers to being able to use AI pervasively."
Via @a16z
$AMZN
Daniel Newman retweeted
Interesting. Who are some of the companies that own scarce compute AND have the necessary power? 👏🏻
Palo Alto Networks CEO Nikesh Arora says the AI profit pool could shift from chips toward owners of scarce compute, because demand is "insatiable"
“So that’s probably what’s going to happen because it’s a matter of the profit pool, right? It all depends on where you shift the profit pool.
“Today, they all live in Nvidia and the memory industry, and they’re being shifted towards the hyperscalers, people who have large compute available, because at one point it’ll become less about the chips, it’ll become more about the availability of compute, because I think demand is insatiable.”
“There’s enough demand. There’s more demand than you and I can imagine. So yes, I understand the enterprise diffusion comment, but it’s going to get diffused.
“The question is, even if you try, if every company in the world today said, I’m in, I’m in, I’m ready, I’m embracing AI, let’s go as fast as we can, there’s not enough compute in the world to satisfy us.”
What does this guy know anyway? A bunch of anon accounts on X are saying no one is using AI and it is game over. 🥴😆
More data showing AI compute is becoming more valuable, not less. 👀
GPUs are aging like fine wine, with rental rates up across the board since January 1st
Approximate change YTD:
- H200: +160%
- B200: +120%
- RTX PRO 6000: +65%
- H100: +35%
Even the 6-year-old A100 rents for more than it did at the start of the year
Charts of the Week: a16z.news/p/charts-of-the-we…
For any of my fellow lifters looking for a great way to build width and activate lats. This one is for you. Absolute 🔥
Have a great Saturday, y’all. 🙌🏻💪🏻
More signal that tech earnings remain strong.
However, it doesn’t mean the market will improve anytime soon. 👀
Daniel Newman retweeted
I am always ‘impressed’ by analysts that make these kinds of calls after the move and sentiment has shifted.
$CRWV “was” overvalued. Now it’s in value territory. When it’s back at $80-$100 the same analysts will swing bullish. Laughable.
BERNSTEIN: “.. we caught a lot of hate when we initiated $CRWV with a $56 price target seven months ago. Today, it’s a lot harder to find bulls — certainly some of them just don’t call us anymore.”
Reiterate Underweight
Interesting. Who are some of the companies that own scarce compute AND have the necessary power? 👏🏻
Palo Alto Networks CEO Nikesh Arora says the AI profit pool could shift from chips toward owners of scarce compute, because demand is "insatiable"
“So that’s probably what’s going to happen because it’s a matter of the profit pool, right? It all depends on where you shift the profit pool.
“Today, they all live in Nvidia and the memory industry, and they’re being shifted towards the hyperscalers, people who have large compute available, because at one point it’ll become less about the chips, it’ll become more about the availability of compute, because I think demand is insatiable.”
“There’s enough demand. There’s more demand than you and I can imagine. So yes, I understand the enterprise diffusion comment, but it’s going to get diffused.
“The question is, even if you try, if every company in the world today said, I’m in, I’m in, I’m ready, I’m embracing AI, let’s go as fast as we can, there’s not enough compute in the world to satisfy us.”
Yes, the path is quite clear.
Lower inference cost -> Exponential usage -> infrastructure growth -> increased worker productivity -> economic (GDP) growth. 💡
CEO of Amazon, Andy Jassy, on the cost of inference:
"The cost of inference has come down quite a bit in the last few years, but it is still an order of magnitude too expensive, in my opinion, for it to be used as pervasively as possible. If you are asking enterprises to rewrite their applications that use CPU today with AI accelerators, if the difference in those costs are very significant. The amount of incremental benefit they have to get from their AI accelerators on there is a much higher bar to want to rewrite those apps."
"If you can take the cost of inference down, that bar is much lower and all applications are going to be better and smarter and more capable with AI accelerators and inference. That is a really important part for customers to being able to use AI pervasively."
Via @a16z
$AMZN
