I doubt we’ll see large scale data center deployments in space in the next five years.
There are too many constraints, particularly related to cooling.
However, just as we’ve seen with quantum stocks ( $IONQ, $QBTS, $RGTI), the market can be very forward looking when an industry is promising enough.

This is why knowing what you invest in is very important
Pre-revenue data center stocks without any proven track record are simply too risky

$IREN is the best at securing power
$CIFR is the best at managing energy resources
$WULF is the best at securing opportunities
$NBIS is the best at software
$CORZ has the most diversified portfolio
$BITF has the most 2-year growth potential
$SLNH is the best microcap bet
Do you agree?

The West is just starting to talk about space data centers
China, however, started its first constellation in May this year
China has already launched the first 12 satellites of a planned 2,800-satellite space computing constellation
This constellation is designed to enable real-time, in-orbit data processing, accelerating AI development in space

People can’t hold a balanced opinion
Neoclouds are the new $NVDA after TPUs
Supply will overwhelmingly outpace demand, and the tokenomics will eventually force them all into bankruptcy
$NBIS, $CRWV, and $IREN will all go to 0, and GPUs will sell for one-tenth of today’s price per hour

Vultr to invest $1 billion in Ohio AI cluster using $AMD chips
The new cluster is a 50-megawatt, 24,000-chip facility that will deploy $AMD's Instinct MI355X
The company forecasts the new cluster to go online by early 2026. "We expect the cluster to be sold before it goes live. We also have MI355X GPUs racked in our Chicago region for customer testing and pilots," Kardwell told Reuters in an interview
In December 2024, Vultr secured a $3.5 billion valuation in a financing round led by LuminArx Capital Management and AMD Ventures, the investment arm of $AMD

From the FT article:
"building new gas capacity now costs about 2,400 dollars per kilowatt, up 71 percent in four years according to Wood Mackenzie"
Energy is rapidly becoming the most precious asset in the AI supercycle
$IREN $CIFR $WULF $CLSK $RIOT $DGXX $BTDR $BITF $GLXY

According to S&P Global, data center demand will grow 56 percent faster than data center supply
That’s insanely bullish for all data center companies
Here’s a list of names:
Large Cap
$EQIX | Equinix - MC: 72.8B
$DLR | Digital Realty Trust - MC: 57.3B
$CRWV | CoreWeave - MC: 44.0B
$NBIS | Nebius Group N.V. - MC: 24.6B
$IREN | IREN - MC: 12.9B
$GLXY | Galaxy Digital - MC: 10.0B
⸻
Mid Cap
$FRMI | Fermi - MC: 9.3B
$APLD | Applied Digital - MC: 8.7B
$CIFR | Cipher Mining - MC: 7.8B
$WULF | TeraWulf - MC: 6.1B
$RIOT | Riot Platforms - MC: 5.6B
$CORZ | Core Scientific - MC: 5.5B
$HUT | Hut 8 - MC: 4

I’d say $IREN could realistically add $20b to its market cap with Sweetwater
If they secure a 2 GW deal with a hyperscaler, the stock is never returning to double digits again
However, I believe they’d be better off focusing on colocation, not cloud
The long-term usefulness of GPUs at this stage is uncertain
We can’t expect the past to repeat itself
AI accelerator development is accelerating, and we have new generations every year
I’d say $IREN would benefit from sticking to their bread and butter, which in my view would provide shareholder returns with much less risk

Optics will likely follow memory in the AI supercycle
Here are some of the top optics stocks to consider for your portfolio:
— $COHR | Coherent – Makes the high-speed lasers and fiber-optic parts that move data between servers in cloud and AI data centers.
— $LITE | Lumentum – Supplies optical modules and lasers that big cloud providers use to upgrade their data-center and AI networks to higher-speed connections.
— $AAOI | Applied Optoelectronics – Builds fiber-optic modules that plug into servers and switches so hyperscalers can move data quickly inside modern data centers.
— $FN | Fabrinet –

$AMD formed a joint venture with Humain and $CSCO to deliver up to 1 GW of AI infrastructure by 2030, starting with a 100 MW deployment in the Kingdom of Saudi Arabia
The project is expected to begin operations in 2026, with a first-phase deployment of 100 MW powered by Humain data center capacity, $AMD Instinct MI450 GPUs, and critical infrastructure solutions from $CSCO

$AMD chief says company ready to pay 15% tax on AI chip shipments to China

$AMD
Lisa Su believes there’s strong growth ahead in CPUs with AI and agentic workloads.
“There’s more work being done, and with things like agent workloads starting, they’re spawning more general-purpose CPU needs.
Because if you think about it, if you have, let’s call it 1,000 agents or 1,000 virtual employees, they need to operate on some data set. They need to operate on some computing capability.
And that requires general-purpose CPUs. So we actually have a view that the CPU market will substantially grow over the next four or five years as we see AI usage really spawn more traditional

$AMD
Lisa Su says they collaborated for 2 years with OpenAI before their industry-shaking deal

$AMD Lisa Su gets asked if she sees a bubble in AI:
"We don't see a bubble.
What we do see is very well-capitalized companies, companies that have significant resources using those resources at this point in time because it's such a special point in time in terms of AI learning and AI capabilities."