South Platte said:

---

flashplayer said:

---

Against my own advice, bought my first shares today on the drop. Plan to wait out some of the lockup to get some more closer to $100.

---

I've love to be a fly on the wall when you sit down and give yourself advice.

---

"Self, I told you not to buy at $189/share"

Space-based compute would be to solve for speed of deployment, and power specifically. The latter is the main challenge it seeks to resolve as a constraint. The prospect of Terafab becoming the 'most valuable building on earth' is intriguing but it will certainly face real challenges.

Quote:

---

What Terafab's Numbers Have Said and When
The investment figure history is worth holding alongside any current announcement:

• March 2026 announcement: $20$25 billion described as initial cost; no binding commitment from any party
• May 2026 regulatory filing: $55 billion stated as proposed initial investment for the Texas facility; $119 billion cited as potential across all phases
• May 2026 SpaceX S-1: Project described as "general framework" with no financial terms, no IP rights, no binding commitments; both Intel and Tesla explicitly noted as "not obligated to remain"
• August 4, 2026 (SpaceX Q2 earnings call): $55 billion cited again as first-phase investment; first-chip production target stated as late 2027
• August 6, 2026 (Governor Abbott announcement): $16.8 billion declared as "first phase," with "future expansion phases bringing total investment much higher"

Electrek, which has tracked the project closely, noted the August 6 figure diverged sharply from Musk's March pitch and represents "roughly a third of the first-phase number the company itself submitted three months ago." SpaceX did not explicitly reconcile the $55B and $16.8B figures in its August 6 communications. The most likely interpretation is that $16.8B represents an initial construction and equipment commitment for the first buildout stage, while $55B encompasses a broader first-phase scope including equipment procurement at full scale but neither company has confirmed that reading publicly.

---

From an investor's perspective, this is a JV between SPCX and TSLA so some of this can be reconciled as such for a fabrication (not a data center) perspective, but the cash flow/capital investment required will be something a lot of folks keep an eye on.

Full disclosure; I hold a few shares but don't really see this as a big stock to own right now, though I am very curious on 'dips' what happens.

Power would be the primary reason for the difference in speed to deploy.


My expectation is that chip manufacturing is going to exceed data center capacity to deploy them (especially if Terafab is successful). That orbital compute will be less profitable on a per GPU basis than terrestrial compute in the immediacy, but the volume capable of being deployed to orbit will be greater than the rate of deployment terrestrially creating more gross profit being generated in orbit than on Earth over time.

That over time the allowance for data center construction will reduce as people recognize that compute can be deployed in greater volumes in orbit without land, water, and electrical use.

That the reduced construction of data centers terrestrial will put a pinch on GPU longevity terrestrially that won't exist in orbit. In other words in 6 years when a new rack is producing 500-1000x tokens per MW relative to Vera Rubin the capital efficient process might be to rip GPUs out of a data center every 5 years while orbital compute will only decommission satellites based upon their actual remaining utility, and their utility over time due to model improvements will actually improve rather than reduce.

Two years between Blackwell and Vera Rubin, 10x productivity per MW

10x more tokens per megawatt.

CoreWeave has the first measured performance of NVIDIA Vera Rubin NVL72, showing 10x improvement in tokens per second per megawatt on DeepSeek-R1 compared to Blackwell. https://t.co/Jz4CwXcOp1

— NVIDIA (@nvidia) July 21, 2026

Delivering a massive upgrade to the world’s #AI infrastructure, our CEO Jensen Huang introduced the NVIDIA Blackwell platform to unleash real-time generative AI on trillion-parameter LLMs at today's #GTC24 keynote. Read more about our announcements. https://t.co/nDVFy00odb

— NVIDIA (@nvidia) March 18, 2024

Elon announced they are going to use their orbital compute rack design for terrestrial applications. I could easily see them deciding to rip racks out of a data center on Earth every 5 years and launching them into space to extract the last remnants of their usefulness while others are just eating the expense of replacement and/or selling their used compute to SpaceX to be redeployed to space for a second life.

The same Starmind V1 satellite design (minus solar & radiator) will be deployed on the ground in our data centers. It’s a major improvement in data center efficiency. https://t.co/0D0ZQUqS4M

— Elon Musk (@elonmusk) August 4, 2026

Both SpaceX (Bastrop, TX) and Tesla (Brookshire, TX) are constructing solar manufacturing. Targeting 100 GW production annually each. SpaceX for Starlink/Starmind and Tesla for Robotaxi/Grid/Megapack storage. Tesla is supposedly integrating edge compute into their Megapack products to provide incremental inference, same is potentially going to be done in Robotaxi and maybe Optimus. Would not be surprised to see them do the same for Powerwall and provide compute subsidized whole home energy back up.

Elon Musk stated that Tesla and SpaceX are both separately working to build up to 100 GW per year of manufactured solar power in the US in about 3 years.

“100mi x 100mi of solar is enough to power the entire US.” pic.twitter.com/9dP2WfcWZp

— Nic Cruz Patane (@niccruzpatane) January 23, 2026

hph6203 said:

---

Power would be the primary reason for the difference in speed to deploy.


My expectation is that chip manufacturing is going to exceed data center capacity to deploy them (especially if Terafab is successful). That orbital compute will be less profitable on a per GPU basis than terrestrial compute in the immediacy, but the volume capable of being deployed to orbit will be greater than the rate of deployment terrestrially creating more gross profit being generated in orbit than on Earth over time.

That over time the allowance for data center construction will reduce as people recognize that compute can be deployed in greater volumes in orbit without land, water, and electrical use.

That the reduced construction of data centers terrestrial will put a pinch on GPU longevity terrestrially that won't exist in orbit. In other words in 6 years when a new rack is producing 500-1000x tokens per MW relative to Vera Rubin the capital efficient process might be to rip GPUs out of a data center every 5 years while orbital compute will only decommission satellites based upon their actual remaining utility, and their utility over time due to model improvements will actually improve rather than reduce.

---

Why? (re the bolded part). Deployment cost will likely be a wash, with launch costs cancelling out terrestrial siting and construction costs. The operating costs will be dramatically lower for orbital because the power is free (solar), the cooling is free (no water fees), no property/operating taxes, no personnel needed onsite to operate, etc. The long terms costs to operate orbital are just a fraction of what they are for terrestrial and the orbital launch costs will be going down over time while the terrestrial siting and construction costs will keep going up.

South Platte said:

---

flashplayer said:

---

Against my own advice, bought my first shares today on the drop. Plan to wait out some of the lockup to get some more closer to $100.

---

I've love to be a fly on the wall when you sit down and give yourself advice.

---

Good thing I listened to myself.

Think you're underestimating hardware and launch costs. I think it will eventually be absolutely cheaper to deploy, but that's probably not by 2030. SpaceX still has quite a way to go to cost down/prove Starship reusability and scale up radiator construction to drag costs down.

hph6203 said:

---

Think you're underestimating hardware and launch costs. I think it will eventually be absolutely cheaper to deploy, but that's probably not by 2030. SpaceX still has quite a way to go to cost down/prove Starship reusability and scale up radiator construction to drag costs down.

---

Right now, people are throwing out these wild estimated for $80-170B per GW for an orbital data center based on non-reusable developmental starship launch costs. When you look at forward projections of rapidly reusable launch costs even by 2028-2029, the cost for an orbital data center drops closer to $20B. Meanwhile, the comparisons are all using outdated land-based numbers that assume $16-20B per GW, when real-world development costs are surging close to $30-40B per GW.

So yes, today, the costs are way higher for orbital...but in the VERY near future, the costs will be a wash or even favorable to orbital, and the long-term operational costs for orbital are a fraction of what it will cost to operate a land-based data center.

txags92 said:

---

hph6203 said:

---

Think you're underestimating hardware and launch costs. I think it will eventually be absolutely cheaper to deploy, but that's probably not by 2030. SpaceX still has quite a way to go to cost down/prove Starship reusability and scale up radiator construction to drag costs down.

---

Right now, people are throwing out these wild estimated for $80-170B per GW for an orbital data center based on non-reusable developmental starship launch costs. When you look at forward projections of rapidly reusable launch costs even by 2028-2029, the cost for an orbital data center drops closer to $20B. Meanwhile, the comparisons are all using outdated land-based numbers that assume $16-20B per GW, when real-world development costs are surging close to $30-40B per GW.

So yes, today, the costs are way higher for orbital...but in the VERY near future, the costs will be a wash or even favorable to orbital, and the long-term operational costs for orbital are a fraction of what it will cost to operate a land-based data center.

---

*Assuming all sorts of hopium

It doesn't matter what the theoretical cost differential is between building data centers on the ground vs in orbit if you can't actually build them on the ground.

I paused new data center projects pending a statewide grid audit.

Data centers must disclose their plans to pay their own way, provide their own power, reuse their own water, reduce the cost of electricity, & avoid disturbing neighborhoods.https://t.co/9n4zMSViep #FoxNews

— Greg Abbott (@GregAbbott_TX) August 5, 2026

Isn't this just optics though? I don't know a single data center pausing construction. Something tells me anything with an approval (no matter phase of construction) is not impacted. That's probably enough projects for 3-4 more years of buildout.

If ERCOT interconnects will be withheld while a grid audit is performed it'll be more than optics. That will delay projects currently in the works and push some projects in planning out of state or beyond their feasibility.

Anything that already got approval won't really have impact, and it's a biiiiiig backlog.

But DCs moving forward wanting to get approval are going to be up against it, at least for a bit.

As soon as AI turns the corner and produces the "killer app" for the common person, much of this protesting will die down.

I'd bet $100 this impacts nothing for 2-3 years. My guess is he got a thumbs up after enough approvals to halt new approvals.

Just FYI...
Google "Amazon 7.65 GW." Amazon is in contract with an already permitted NG generation facility that has broken ground in Pecos County. The entire generation plant will be to power an Amazon DC with possible future grind interconnect if feasible.

It doesn't affect Terrafab, the TSLA/SPCX facility in Grimes county at all as it is not going to be connected to the grid. SPCX is around $143, up around 4 percent for the past 30 days.

Dumb question, with the Tesla CyberCab including an integrated Starlink satellite antenna and eventually each Tesla sold will have an integrated antenna, does SpaceX automatically count those as new Starlink subscribers on the public quarterly/annual reports?

Quote:

---

"Starlink Subscriber" refers to a unique Service Line that is directly assigned to a Starlink.com account registered to a person or entity that does not have a direct, negotiated agreement with the Starlink sales team.

---

CyberCab and Tesla vehicles would fall under a different category. Even if Tesla and SpaceX merge they would still not be accounted for as individual subscribers, because they would be under separate business units.

Depends on how they handle the service when they sell the car. If it is included in the purchase price, then I would assume no. If they come included with the starlink receiver, but the customers have to pay a monthly fee to use it, I would assume anybody paying the fee would count as a subscriber.

Tesla would be the entity contracted with SpaceX for bandwidth and then Tesla resells the subscription to individual car owners. They would be counted as individual subscribers for Tesla Premium Connectivity subscription plan, and as a corporate contract/not a subscriber for Starlink's disclosures. What I quoted above is the definition of subscriber from SpaceX's SEC filings.



For the data centers won't be put into orbit crowd:

My directive has halted up to 1,800 data center projects.

I established guidelines & guardrails to protect Texas communities:

✔️Data centers must not take water needed by local communities,
✔️They must not take power needed by the Texas power grid,
✔️They must lower the cost…

— Greg Abbott (@GregAbbott_TX) August 20, 2026

The National Republican Senatorial Committee sent a private memo yesterday to US AI companies warning them that the GOP is on the verge of losing Ohio over data centers, and unless something is done to improve public perception, political support will collapse nationwide. pic.twitter.com/WWyz9QgG2u

— Andrew Curran (@AndrewCurran_) August 20, 2026

We’re not going to let greedy developers bully their way into our communities in order to build reckless AI data centers.

If you want to build here in Pennsylvania, you’re going to need to follow the strictest standards in the nation — or you better find somewhere else. pic.twitter.com/IMd5OSeZ1R

— Governor Josh Shapiro (@GovernorShapiro) August 19, 2026

More.

What a year it’s been https://t.co/UTJM4cN8nZ pic.twitter.com/v18krjouYA

— Matthew Yglesias (@mattyglesias) August 20, 2026

Agreed. The anti-data center movement is mind boggling. This is not a race we can afford to lose and Elon isn't ready to start launching them to space. This is as nuts as people complaining about railroads in the 1850s, electricity in the 1880s, and the production of oil in the early 1900s. So many Karens, and so many companies that need to step up their efforts to assuage whatever concerns these communities have.

Yeah, I'll raise my hand here. Local governments have done a sh8t job. I'm not very NIMBY, but this rollout has been atrocious. I don't even have a Facebook or whatever, so I'm not sure what Chicom psyop is targeting me. Believe it or not, most people see an absolute mismatch in the Black Rock and FAANG/MANGO-led developers and the (I'll be gentle here) unsophisticated local leadership. None of it should have been done with such a lack of transparency. But that's the way the developers wanted it, and they got it. That's the way conservative politicians wanted it, and they got it. My small town is led by a guy who sells insurance to local companies (accounting degree), and they have built 3+ with another 4-5 to go (no one knows the real number). I'm sorry, but he might be a great guy, but I don't trust his sole judgement for a minute in this arena.

Don't mistake what I am saying. I am blaming both sides. They need to come together and meet each other's needs so we don't start losing this arms race.

Oh, my mistake.

I guess you're blaming both sides, saying one side is mind boggling, same side is a bunch of Karens, and same side is like blocking electricity.

The other side just needs to explain it all away through.

Let me explain this part further:

Quote:

---

and so many companies that need to step up their efforts to assuage whatever concerns these communities have.

---

I think some of the designs are irresponsible by the developers. That does need to be addressed. As you said, the local authorities are failing there. That doesn't mean I think those institutions are blameless either. They have to use common damn sense and can't just hoard utilities for themselves. There are more sustainable ways they can build and still walk away with a decent profit margin and a community that benefits. It is not a zero sum game. But they got greedy.

Same thing happened at the beginning of all those historic booms.

The problem is people aren't just saying not in my back yard. They are starting to say not in anyone's backyard. Start learning chinese then.

Agreed, this didn't have to be like this.

My parents are in rural Falls county and at the county line of Milam and Falls counties, OpenAI is building a data center there.

Trucks are taking water out of the town lake, and no one in that town seems to know a thing about it. I guess said small town government agreed to terms with OpenAI to provide temporary water for construction.

Yeah, my problem is that there are lots of issues that need to be adequately addressed by the data center developers to make people happy. And all of those issues can be addressed by their local governments as part of whatever property tax abatement deal they work out with the developer. But it has already been documented that far left groups whose funding is tied directly to China are propping up fake "grassroots" opposition movements that are duping lots of rural folks to join them, and their goal is to stop all data center construction so that China has a chance to catch up to us.

The groups pushing for statewide bans or moratoriums on data center construction can mostly be tied back directly to those Chinese funded astroturfing efforts, and their goal is not to address local rural concerns about data center construction. Their goal is to tie up all data center construction with red tape while it moves full speed ahead in China.

The tools to address the concerns about data centers are already in the hands of the local cities and counties. These data centers are looking for property tax abatements in return for the millions of dollars of other tax revenue they bring to areas they are built. If you want dark skies lighting? Put it in the tax abatement agreement. If you want them to use a closed circuit cooling system to minimize water consumption, put it in the agreement. If you want them to produce their own power so they don't overtax the power grid, put it in the agreement. But if you do demand they generate their own power, don't turn around and protest the natural gas fired power plant they want to build next door.

If you just want them not to build next door to you, then band together with your neighbors and go buy the property next door so they don't build there. Otherwise, this is Texas and if they buy the property and satisfy all of the laws and agreements about what they need to do, then they have a right to build something there.

The biggest mistake in all of this is looking for the state level government to step in and fix this. If they do, you won't like the way it happens. At the state level, the data center lobbyists will make sure that whatever state level solution gets crafted is very friendly to them and imposes the least amount of restrictions. The state level politicians can use the money the data center builders throw their way to overcome any static that comes their way from Podunk County in the next election.

You have the most influence over your local government. They know that no matter how much money the data center builder throws their way, that if they piss off the local public, they will be voted out of office in the next election. So put the pressure on the local governments to negotiate the things that are important to the local setting as part of the development agreements. Stop looking to the state or federal governments to fix what should be local issues. And stop using NIMBY to try to make them go do it in somebody else's yard. That is how so much of our manufacturing ended up in China to begin with, because we made it too hard to do it here.

500,000ags said:

---

Agreed, this didn't have to be like this.

My parents are in rural Falls county and at the county line of Milam and Falls counties, OpenAI is building a data center there.

Trucks are taking water out of the town lake, and no one in that town seems to know a thing about it. I guess said small town government agreed to terms with OpenAI to provide temporary water for construction.

---

How is that different from any of a thousand other construction projects that work out deals with local entities to pull water from fire hydrants or ponds or creeks for dust control or anything else? Why is it a big deal when they do it for construction of a data center? The town has the ability to change where they get their water if it is a big deal. That can be fixed at the town level and doesn't need Greg Abbott's input.

Because the data center isn't in the town or county even. Much like with most of this, no one really knows how the town, or its resources, is impacted (positively or negatively). We will have to just disagree that the state's intervention should already be local enough to address this issue. If this were < 10 locales across the entire state, I'd be inclined to agree, but this is dozens and dozens of projects across the entire state. This is a state-wide issue. IMO, most conservatives (which I'm going to assume you are) are too quick to say this is a local issue, when current opposition that is local is just dismissed as "funded by China." Doesn't make sense. There could be some of that, but I know plenty of people on both sides of the isle that are not impacted by that whatsoever.

500,000ags said:

---

Because the data center isn't in the town or county even. Much like with most of this, no one really knows how the town, or its resources, is impacted (positively or negatively). We will have to just disagree that the state's intervention should already be local enough to address this issue. If this were < 10 locales across the entire state, I'd be inclined to agree, but this is dozens and dozens of projects across the entire state. This is a state-wide issue. IMO, most conservatives (which I'm going to assume you are) are too quick to say this is a local issue, when current opposition that is local is just dismissed as "funded by China." Doesn't make sense. There could be some of that, but I know plenty of people on both sides of the isle that are not impacted by that whatsoever.

---

There is not a single data center being built in Texas that is not subject to taxation by a county government. Every square inch of land in Texas is part of a county. Has anybody gone to the county or the city and asked? Or are they all just making posts on Nextdoor saying "why are they pulling water from Lake ___?" and letting tons of misinformation and guessing take the place of actual information that is almost certainly on file with the city or the county and can be obtained by FOIA if a simple conversation with a clerk or engineer can't get it for you.

I am not dismissing the local opposition. I am encouraging that local opposition to focus on getting the things they are concerned about (water use, grid impact, noise pollution, light pollution, etc.) addressed at the local level instead of pushing for a complete stop on all projects. Pushing for a complete ban or a moratorium with no end point is doing China's work for them, whether your group is funded by them or not. Don't fall for the propaganda being funded by the Chinese influence groups. Do your local due diligence, talk to your local governments, and make your voices heard about what is important.

Or don't. Keep pushing for a complete ban and it will make my SpaceX stock worth that much more when Elon starts launching them into space. And the local cities and counties won't get any of the tax revenue they would have received and the local facilities here in Bastrop County and in Grimes County that are going to be making the datacenters going into space will just generate that much more local revenue here.

By your naivety, I do wonder how close these buildouts are to you? There is very little that is readily transparent or makes sense.

My parent's town (and the town's lake) is in Falls county and the data center is in Milam county - only 2 miles away from the lake though. I understand unincorporated areas are in a county. These are small towns that don't use Nextdoor. You're throwing a lot out, so this is piece meal. I believe you are trying to make it sound like people are just working themselves up without grounds. That is not the case IME. My mom asked the city, who sent her to the county, who sent her to the water company, who sent her back to the county. It's never super easy.

I'm not in the habit of falling for propaganda, no matter what country is shilling it. I never said I was for a complete ban, and most people in my city are not either. They are for transparency and detail. In my city, information pulls were initially denied repeatedly regarding data centers. Even local papers were often denied. As the raised stink was too much to ignore, more detail was released, but mainly about water, which I through was addressed somewhat. I am more worried about backup power emissions at the planned level of development (and close proximity) and grid reliability, which the city has released nothing about.

Data centers aren't the devil, but like I said previously, the process has been atrocious.

500,000ags said:

---

By your naivety, I do wonder how close these buildouts are to you? There is very little that is readily transparent or makes sense.

My parent's town (and the town's lake) is in Falls county and the data center is in Milam county - only 2 miles away from the lake though. I understand unincorporated areas are in a county. These are small towns that don't use Nextdoor. You're throwing a lot out, so this is piece meal. I believe you are trying to make it sound like people are just working themselves up without grounds. That is not the case IME. My mom asked the city, who sent her to the county, who sent her to the water company, who sent her back to the county. It's never super easy.

I'm not in the habit of falling for propaganda, no matter what country is shilling it. I never said I was for a complete ban, and most people in my city are not either. They are for transparency and detail. In my city, information pulls were initially denied repeatedly regarding data centers. Even local papers were often denied. As the raised stink was too much to ignore, more detail was released, but mainly about water, which I through was addressed somewhat. I am more worried about backup power emissions at the planned level of development (and close proximity) and grid reliability, which the city has released nothing about.

Data centers aren't the devil, but like I said previously, the process has been atrocious.

---

I live in the county about 2 miles as the crow flies from where Elon is building his current Starlink consumer terminals and where he will be adding hundreds of thousands of square feet to manufacture his data center satellites. There are a number of data centers in various stages of being built nearby in Bastrop County. There are locals here who have gone to the county engineer and other county leaders and obtained copies of the development agreements. It is not hard to do. All of the documents are FOIA eligible, so they have to give them to you if you ask formally although they can redact proprietary info. The meetings where the agreements are voted on are subject to the Texas Open Meetings act and anybody can attend.

Would it be easier if the developers held open houses to talk about their plans? Yes. And in many cases they are starting to do that now, recognizing that gaining local trust is an important part of ensuring that approval happens with the local governments. But they also are not going to tell anybody their plans before the land is purchased and the deals are in place because disclosing info too early can cost them millions in extra costs or could even scuttle the deal altogether.

And just FYI, Nextdoor is everywhere. Just because you aren't on it doesn't mean your neighbors aren't. And I see all kinds of misinformation and NIMBY crap going on there by people who are all riled up by false info and who ignore the folks posting actual copies of the agreements and actual info obtained from the County about the various sites. Those kinds of people are the ones who are most susceptible to being pushed by outside influences like China whose interests are in shutting it all down.

My apologies if I lumped you in with others who are asking for a complete ban or open ended moratorium, but that seems to be the big push happening right now. Places like Hays county and others have already put them in place with no clear information about what exactly they are waiting for or what conditions would cause them to lift them. Many of those groups are pushing hard for a state level regulatory scheme and my opinion is that anything that comes from the state down will be a one-size fits all solution that is not responsive to specific local concerns. And in general, it will be very friendly to the data center developers just because it is easier for their money to control politicians at the state level.

Don't really see a problem with state wide minimum requirements for data center builds. I don't have much faith in most local governments to navigate those base level requirements and this is a gold rush that will have all kinds trying to exploit opportunities, and you don't want a slew of exploitative and/or half developed projects that failed prior to completion.

Concern is moreso that it becomes politically necessary to be anti-data center as a blanket policy which is the direction things are trending.



People see data centers as a threat to their water, their electric bill and their job and the positive case has not been well argued by the AI lab leaders. They do a lot of talking about risks that they absolutely should not do. There's plenty of voices for the dangers already. It's their job internally to mitigate risks and their job externally to explain benefits. Don't talk about doctors losing their jobs, talk about healthcare so cheap it will be like buying a burger and fries and that the burger and fries will feel more affordable as a consequence.

ELON MUSK: "If you've got humanoid robots that have very high dexterity and are incredibly smart, it means that everyone on earth will have access to better medical care, than the richest person on Earth.

I had to have like a neck surgery three times because the first two ones… pic.twitter.com/CD2BuT6b3Z

— DogeDesigner (@cb_doge) August 20, 2026

hph6203 said:

---

Don't really see a problem with state wide minimum requirements for data center builds. I don't have much faith in most local governments to navigate those base level requirements and this is a gold rush that will have all kinds trying to exploit opportunities, and you don't want a slew of exploitative and/or half developed projects that failed prior to completion.

Concern is moreso that it becomes politically necessary to be anti-data center as a blanket policy which is the direction things are trending.



People see data centers as a threat to their water, their electric bill and their job and the positive case has not been well argued by the AI lab leaders. They do a lot of talking about risks that they absolutely should not do. There's plenty of voices for the dangers already. It's their job internally to mitigate risks and their job externally to explain benefits. Don't talk about doctors losing their jobs, talk about healthcare so cheap it will be like buying a burger and fries and that the burger and fries will feel more affordable as a consequence.

ELON MUSK: "If you've got humanoid robots that have very high dexterity and are incredibly smart, it means that everyone on earth will have access to better medical care, than the richest person on Earth.

I had to have like a neck surgery three times because the first two ones… pic.twitter.com/CD2BuT6b3Z

— DogeDesigner (@cb_doge) August 20, 2026

---

You and I will have to agree to disagree on whether state wide requirements will be more or less responsive to local issues and concerns or not. Local politicians are very susceptible to smaller groups of people willing to work against them if they don't do their job. It takes a lot of money to unseat an incumbent state legislator or statewide official and the data centers will always have more money to spend buying state level offices. You will always have more power to influence your local officials, and that starts with reaching out and talking to them or their staff.

I agree with you that somebody needs to be arguing the "for" case more convincingly, and I don't think that is really going to be best suited coming from individual project developers. The people who need to hear the benefits the most aren't going to listen to it if it comes from the company wanting to build it.

Anyhow, that is probably enough of a derail of this thread on my part. Would be glad to continue via PM or in another thread if you feel the desire.