Tesla investor since 2016, long term fundamentals 🇺🇸 DON’T PANIC
United States
Joined December 2021
- Tweets7.5K
- Following3.2K
- Followers3.8K
- Likes61.6K
Peter retweeted
Cybercabs now picking people up from the F1 event.
Peter retweeted
Like I mentioned earlier, this is the end game. $SPCX will become the biggest hyperscaler within 2 years or so.
Having top rocket engineers to build data centers is unfair, I know, I know, but you know what? I love the unfair competition. And I am not even talking about the ODC yet.
The king is in the house. And this is the end game.
On SpaceX’s $SPCX first earnings call, Elon said it plainly: “We decided to go all-in on NVIDIA because the Vera Rubin architecture is the best. We are exclusive to Nvidia $NVDA .”
The space version is Starmind AI1: one satellite = one full Vera Rubin NVL72 rack (72 Rubin GPUs + 36 Vera CPUs, ~3,600 PFLOPS NVFP4, 20.7 TB HBM4, 120–150 kW). Cooling radiates into vacuum, solar is free. Prototypes early 2027, deployment Q4 2027, scale 2028. Built at Gigasat, Bastrop.
NVIDIA’s Space-1 Vera Rubin Module claims 25× the space inference of an H100. Jensen called space data centers with SpaceX “a particularly new market.”
On the ground: ~200k GB300s online per month. Next year shifts to Vera Rubin. Targets: 440k-GPU training cluster before Q2 2027, 10 GW incremental buildout in 2027.
Elon warned of stranded compute in half-built data centers. GPUs bought, power not connected. The “lock early, resell at 2–3×” model fails when the bottleneck is power, grid, cooling, and permitting — not the chip.
Ground centers hit power, cooling, and grid walls together. In orbit, sunlight is free and heat radiates into space. Earth isn’t out of room — it’s out of immediately usable room. Space is one of the fastest paths.
Superintelligence doesn’t start with parameter count. It starts when compute demand and supply enter a verified positive feedback loop — and that loop is only just beginning.
Peter retweeted
I sell data center capacity for a living. The conversation with customers has changed completely in the last eighteen months.
It used to start with price per kW. Now the first question out of everyone’s mouth is when can you actually give me power, and the honest answer in most major metros is later than you want to hear.
Everyone wants to argue about GPU supply. That is not where the wall is. The wall is a transformer you order today that arrives in three years, a permitting process that answers to a county board, and an interconnection queue with roughly 2,000 gigawatts stacked up in it. PJM ran the numbers on projects that came online last year and the average was over seven years from request to live power. Three years to get an agreement, four more years to actually energize after the agreement was signed.
And this is the easy part. We are nowhere near the peak of this demand curve. There is at least another order of magnitude coming, and the grid that is supposed to absorb it operates on utility capital planning cycles that were designed in a world where load growth was flat for twenty years.
Which is why the orbital compute argument stopped sounding exotic to me sometime around last fall. Up there you get uninterrupted sunlight, you dump heat straight into vacuum instead of running chillers, and there is no zoning hearing, no queue position, no substation to wait on.
Plenty of companies have interesting orbital compute concepts. Only one of them owns the rocket, has the scale, and deeply embedded partnership with Nvidia.
SpaceX filed with the FCC in January to launch and operate up to a million compute satellites, and there is exactly one launch provider on Earth with the cadence and the cost structure to make that number anything other than a press release. Starship is the reason this is an infrastructure plan and not a science project.
The constraint on the ground is the grid. The constraint in orbit is mass to LEO. SpaceX is the only company that has solved the second one.
$SPCX
Nvidia today announced that @spacexai will deploy Vera CPUs to accelerate its next generation of agentic AI applications.
"SpaceXAI plans to expand its AI infrastructure behind Grok on Nvidia Vera Rubin, while extending an optimized Vera Rubin NVL72 into space with its first-generation Starmind satellite.
SpaceXAI’s work with Nvidia is moving beyond AI factories on Earth. SpaceXAI is developing AI computing infrastructure for orbit, where power, thermal management, bandwidth, reliability and physical integration impose dramatically different constraints from conventional data centers.
SpaceXAI’s planned first-generation Starmind AI satellite will be based on the optimized Nvidia Vera Rubin NVL72 rack-scale system, extending the same accelerated computing architecture powering next-generation AI factories on Earth into space.
Nvidia and SpaceXAI are working to adapt that foundation to the requirements of orbital computing while preserving a common Nvidia architecture and software ecosystem.
This delivers one computing foundation across a wide range of environments: Vera CPUs accelerating increasingly sophisticated AI agents, Vera Rubin powering the AI infrastructure behind Grok and gigawatt-scale AI factories on Earth, and Nvidia accelerated computing extending into orbital AI infrastructure."
Peter retweeted
Now listen carefully! Here’s one of the most important things in $SPCX Q3 earnings call most investors didn’t pay attention to.
Elon on the Q3 call, almost in passing:
“We already are in early deployment with some trusted customers, and we expect to be in wide deployment within a few weeks for enterprise customers for Grok Bot.”
The numbers behind that sentence:
• Grok Bot launched August. 800K paid weekly users by early October. Doubled in one month.
• B2B is the fastest-growing category. Companies using it across sales, marketing, IT, engineering — as an always-on teammate, not a chatbot.
• Enterprise customers with $1M+ run-rate revenue, including Cursor, up 50% QoQ pro forma.
• New bundled plan: Grok + Grok Bot + Cursor + X Premium on a single subscription.
• Enterprise tier already went live September 4. Two weeks free for all Grok and Cursor enterprise customers.
Here’s what most people missed.
The $500B–$1T YE2027 ARR target gets framed as cloud leasing. But Grok Bot is the other leg — internal compute monetized directly through subscriptions, no intermediary, no spot-price negotiation, no 6–12 month payback math. It’s recurring revenue that scales with seats, not gigawatts rented.
If enterprise demand is genuinely queuing up, that 50/50 internal-external compute split stops being a cost center debate and starts being a revenue allocation decision. Half your fleet earning subscription ARR while the other half earns spot. That’s not a rocket company with an AI side project. That’s a compute monopoly with two monetization engines firing at once.
Peter retweeted
I have friends buying new Teslas because their spouse took their “personal chauffeur” and refuses to give it back.
Peter retweeted
Yes, we are at 11.3 million shares signed.
Freaking unbelievable.
Please keep signing, we need to send this out on Monday, so time is short:
->
surveymonkey.com/r/mergealre…
TESLA/SPACEXSI
MERGE ALREADY!!
The merger is the clear will of Tesla shareholders. In just 24 hours THOUSANDS signed The Joint Shareholder Letter to the Board/Special committee in support of a Tesla/SpaceX merger.
Tesla shareholders are a force:
Together we represent a staggering 11.3 MILLION Tesla shares. Unreal. To put that in context, that is more than HSBC or BNP or Citibank or RBC or Ark.
Let’s keep moving up the list! The letter will close later today to mail to the Board ASAP. Share the link everywhere. SIGN NOW!!! @TeslaBoomerMama
$TSLA $SPCX
surveymonkey.com/r/mergealre…
Peter retweeted
Nick Wray, the 8th person in the world with a Neuralink implant has passed away due to ALS.
For the time I knew him, he carried an optimistic energy that helped pick others up.
One of his go-to sayings was: "Be good, do good. 😎"
In support of his family and friends, we can all choose to live by that message every day.
More information can be found in the link in Audrey's post:
Saddened to learn about the passing of Nicholas "Nick" Wray. Here is his funeral information for anyone who is interested: everloved.com/life-of/nichol…
This shopper @bot will save you money
My shopper @bot template changes the way you buy things. It connects to Amazon, Costco, Instacart, Walmart and it’s easy to add other stores you use often.
You tell it what you want to buy and it will research options to find the best product and then source the best price.
It works with one-of purchases or with regular groceries etc.
The idea is to offload the cognitive load of comparing options, it looks at the number of reviews, average customer review, thinks about what is best for you based on what it knows about you and then hunts for the best price across stores.
After all that research it can present top 2-3 options before going ahead with the purchase for you and will learn your preferences over time so eventually you can trust it to just research and auto-buy for you.
It can connect your credit card with Link plugin or via the default card setup on each account and checkout for you using the grok bot computer.
It has routines to monitor prices and catch when there’s a good deal.
It will send a weekly reminder of things you’ve purchased which are still within the return window so if you change your mind you can return - and will process the return for you so you have the QR code for drop-off to make it easy.
There is a weekly routine that also monitors the price of things you’ve purchased and if they drop in price and still within the return window it will let you know so that you could return and re-purchase at a discount.
The idea is grok bot becomes your interface and manages your shopping accounts so you just have to ask and it does everything for you.
The way I use it is to just use voice mode and say something like:
“I’m looking for a new TV around 55”, I like LG and Samsung but I don’t know which model, needs to be good for gaming and movies, I already have a good sound bar. l
Let me know what the best options are and where to buy”
It will work in the background and ping you when it’s ready.
When you click this share link, it will show the setup I’ve credited across instructions, memories, skills, routines, plugins - so you can take a look before installing it.
x.ai/bot/--X3KeUBk4AwgtfcxxK…
Peter retweeted
I don’t currently know which politician this was, but I don’t make the rules:
It’s all of our duty to share this young man’s rabbit art, after the authoritarian response he received! 🐇 🐰
