Grey Swan Capital Join Now

Article · November 8, 2026

The Servers Can Wait. Gentry’s Neighbors Can’t.

Midterms are slowing the data-center rush. Clarksville shows why Arkansas should ask a different question: who owns the power, and can it keep funding care?

Grey Swan Capital · Article

The Servers Can Wait. Gentry’s Neighbors Can’t.

Midterms are slowing the data-center rush. Clarksville shows why Arkansas should ask a different question: who owns the power, and can it keep funding care?

By Steve Schroeder ·

Originally on Substack: https://greyswancapital.substack.com/p/the-servers-can-wait-gentrys-neighbors

This article includes a video. Watch it in the original on Substack.

It is an example of the kind of teaching tool members can use to understand an issue and start a better local conversation. Look at the work at greyswancapital.ai, and see membership. This is not a fundraiser for a city deal.


Cold Open: Two Buildings. Two Different Questions.

Picture a split screen. This is an illustration, not footage of a particular meeting.

On one side, a county meeting room. A developer has a map, a power requirement, and a request for tax relief.

On the other, a home where people need care every day. No ribbon cutting. Just another shift that has to be staffed.

I live in Conway, Arkansas. My son Gentry lives at the Conway Human Development Center. From where I sit, our HDCs have gone 11 years without the funding they need.

So when somebody says a compute campus could transform a county, I don’t start with how many servers fit inside.

I call them compute campuses because they should be community partnerships and full educational facilities where everyone can win—not private server warehouses. You’ll still see “data center” in news reports and in some project names; that’s their language, not the standard I’m using here.

I start with this: Who owns the power—and what happens to the people who need care after the construction crews leave?

That’s the question underneath this week’s GETM sweep: Midterms Slow AI Servers. God, Energy, Technology, Money. Not four separate drawers. One county has to live with all four.

Scene 1: One Map. Twenty-Two Names. One Deal at a Time.

Aterio’s Arkansas tracker, dated November 8, 2026, lists 34 facilities and about 3,618 megawatts: four operational, six under construction, and 24 announced.

Aterio counts individual buildings or phases as facilities. That does not mean 34 separately named campuses. Four named providers account for 25 of the 34 facilities and 3,155 MW—about 87% of the megawatts: Google, 966 MW / 10 facilities; AVAIO Digital, 957 MW / 7; Clean Cloud Energy, 800 MW / 4; Serverfarm, 432 MW / 4.

That leaves nine slots with no publicly assigned provider. We leave them unmarked. We don’t name them, guess at them, or fill them with somebody else’s list. When we number slots, that’s a way of picturing Aterio’s totals—not recovered Aterio IDs.

Now the lights come up on the cast we can actually name.

Here are 22 named public records of Arkansas projects and operations—22 distinct records, not 22 Aterio slots. Some are big proposals. Some are crypto mines or older colocation sites. Two were cancelled and one application was withdrawn. Don’t subtract 22 from 34 and call the difference a hidden list; campuses hold several facilities, and the two counts measure different things.

  1. Serverfarm ARK (ARK1–ARK4) — Clarksville / Johnson County — under construction; 432 MW IT, 660 MW on-site generation reported.

  2. Google West Memphis / Project Pyramid — West Memphis / Crittenden County — under construction / phased development; 600 MW reported (Aterio’s county estimate is 486).

  3. AVAIO Digital Leo — near Wrightsville / Pulaski County — ground broken, the developer reports; 150 MW initial reported, 1,070 MW planned power.

  4. Sheridan / Clean Cloud Energy — northwest of Sheridan / Grant County — proposed; EdgeConneX due diligence reported; 800 MW campus. Grant County’s 20% abatement is a negotiating starting position, not signed final terms.

  5. Google Port of Little Rock / Willowbend Capital — Little Rock, near Sweet Home / Pulaski County — planned; redesign / permitting; more than 100 MW reported. Sweet Home and the Port are counted once here, not twice.

  6. Conway / Forgelight Ventures LLC — Conway / Faulkner County — proposed; 10 MW reported. The end user is unnamed.

  7. Z Squared / Paradox Data AI-HPC campus — near El Dorado / Union County — proposed expansion, up to 150 MW. Union County has a reported one-year moratorium; the expansion is not approved or under construction.

  8. NewRays One bitcoin mine — Bono community / Faulkner County — operating; MW not verified.

  9. BTC Digital Manila hosting facility — Manila / Mississippi County — operating; additional hosting announced.

  10. Project Pulse — Gum Springs / Clark County — cancelled, tracker reports.

  11. Apollo Project – Helena Harbor — Phillips County — cancelled, tracker reports; 500 MW proposed.

  12. Ritter Data Technology Center — Jonesboro / Craighead County — operating; 0.75 MW in one directory.

  13. TierPoint Little Rock — Little Rock / Pulaski County — operating; MW not verified.

  14. Bentaus / Chadbourn Infrastructure Partners — north of Texarkana / Miller County — announced; approval and build not verified; 200 MW planned.

  15. BlockOps Moreland bitcoin mine — Moreland / Pope County — operating, the operator says.

  16. BlockOps Russellville bitcoin mine — Russellville / Pope County — operating, the operator says.

  17. BlockOps Helena-West Helena — Phillips County — approved 2025; operating per the operator.

  18. Artemis Technologies Cave City — Sharp County — construction begun April 2025; completion unverified.

  19. Artemis Technologies Monticello — Drew County — earlier location reported; current operation unverified.

  20. Artemis Technologies Newark — Independence County — earlier location reported; current operation unverified.

  21. Green Digital / Metahash Global Harrison mine — Boone County — application withdrawn in May 2023, local reporting.

  22. Mainstream Technologies — downtown Little Rock / Pulaski County — operating colocation and hosting.

And one with no name on the door: a second Grant County proposal on Highway 46 near Redfield. October reporting describes signed NDAs, copies released with the developer’s name redacted, and an unnamed end user. We don’t know the developer or the megawatts, and we won’t make them up.

A public directory lists other operating sites as well. Those are directory leads, not verified new hyperscale projects, and they don’t fill Aterio’s missing nine.

The secrecy is the problem

Why are names missing? Two reasons we can document. Aterio sells its full facility-level list behind a paywall. And NDAs, project LLCs, and unnamed end users keep real negotiations out of view—Redfield is the plain example. We can’t prove every missing name is under an NDA.

But here’s my argument, and it’s mine and Grey Swan’s: the secrecy itself is the problem. When a community can’t see who is coming, what was promised, or who will own the power, a hidden private agenda is setting the terms, and the people who live there aren’t partners. They’re spectators. That’s a civic argument about how these deals are made—not an accusation of a crime, a bribe, or misconduct by any named official.

Grey Swan will go down each one, one at a time. Today is Clarksville.

That means the deal the community made, whether the public terms look balanced, how we’d approach it differently, and what choices may still remain. The others get their own examination, not a pretend verdict squeezed into today’s list.

Civic analysis from public sources, not legal advice; based on public reports and Aterio’s public tracker as of November 8, 2026. No wrongdoing by any named official is alleged.

Scene 2: Clarksville Bought Benefits. It Didn’t Buy the Power.

The camera moves to Clarksville, in Johnson County.

Serverfarm’s ARK campus is under construction: four buildings, ARK1 through ARK4, with 432 megawatts of IT capacity. The associated on-site gas plan is 660 megawatts—40 Solaris turbines rated at 16.5 megawatts each. Clarksville Connected Utilities supplies 18 megawatts for construction only. That is not the same as CCU powering the finished campus.

Behind the meter means generation serves the project on its side of the utility meter. It doesn’t mean the town owns it.

The stated investment is $6.6 billion. Under the reported Act 9 bond structure, the city isn’t liable for repayment. That’s an important protection, not a footnote.

The public also gets benefits. The Community Development Grant Agreement provides $1 million each to the city, the county, and Johnson Regional Medical Center, followed by ongoing building payments. I don’t dismiss money going to a local hospital.

But set those benefits beside the other terms: a 65% property-tax abatement for up to 30 years; a Solaris lease of 43 city acres at 6 Graves Lane for $1,000 a month for five years; and no public job guarantee or clawback found in the material reviewed. Permit-fee reports differ: $840,000 in local reporting, $2.1 million on Serverfarm’s site. I’m not going to turn a disagreement into a settled number.

The lease report describes staging and laydown uses. It should not be presented as proof that all 660 megawatts sit on those 43 acres, or that this rent buys access to their output.

Here’s the distinction that matters: the generation belongs on the Serverfarm/Solaris side of the deal, not to the town. The city’s title-holding role in bond financing is not the same thing as public ownership of the generation’s economic upside.

To be fair, Serverfarm does publish planned workforce partnerships with local schools, Arkansas Tech, and the University of the Ozarks. That’s a real potential benefit. The question is whether education and public ownership are written in as partnership terms, or left as goodwill.

That doesn’t make this a scandal. It makes it a deal worth reading.

My fairness reading: there are real public benefits and a real protection against city bond liability. But a long abatement and fixed land rent deserve scrutiny beside private ownership of a productive power asset. Missing public job guarantees or clawbacks are records observations—not proof that the full agreements contain none, and not a charge against anyone.

Sheridan is a useful contrast. Grant County’s negotiating starting position is a 20% abatement, tied to protections on rates, water, internet, noise, and odor. That’s a starting position, not signed final terms or a completed sale. Still, it shows Clarksville’s 65% abatement isn’t the only bargain a county can ask for.

How I’d do it differently: put county-owned or shared generation on the table before incentives are settled. A public authority, county, or qualified partnership might own a meaningful stake in the plant, with transparent financing, operating responsibility, and resident cost protections. After operating costs, reserves, and debt service, a written share of net benefit could go toward care for the vulnerable, including Conway HDC. That’s a proposed alternative, not a term in Clarksville’s deal or a guarantee that public ownership would pay off.

What can Clarksville still do? Ask for the full agreements. Ask how rate protection, water reporting, permanent-job reporting, and decommissioning are handled. At a future lease renewal or a genuinely open amendment, explore fair land terms or a negotiated public stake in generation. Don’t pretend a signed agreement can simply be rewritten, or that later phases remain legally open leverage without checking.

Those are questions for residents and qualified advisers, not a verdict from my desk.

Sources: local agreement reporting, Solaris lease reporting, Serverfarm’s project description.

Scene 3: The Ballot Box Is Now Part of the Power Supply

Cut from a turbine to a voting booth.

The midterms didn’t turn off the internet. They did put another clock on construction.

The Guardian reports that voters approved data-center bans or restrictions in Ohio communities, temporary moratoriums in Maine communities, and a high-impact data-center ban in Emporia, Kansas. It also reports moratorium pledges from newly elected Democratic governors in Georgia and Ohio.

In Texas, the distinction is governor-elect, not a new governor already exercising power. The Texas Tribune reports that Gina Hinojosa has called for a moratorium until lawmakers pass community protections. That is a stated policy direction, not proof that her proposed laws have passed.

Aterio’s national tracker places Texas at the top of its estimated pipeline. Political decisions in a state that large matter beyond its borders. Power approvals, water arrangements, and tax incentives aren’t peripheral to a compute campus. They’re part of whether it can happen.

My reading: treating these only as private server warehouses now carries more political and scheduling risk. That doesn’t mean every project stops, or that every Democrat opposes building. It means a county needs more than a promise that the servers are coming.

Midterms, uncertainty, and negative public opinion are not reasons to freeze. They are ideal times to act on a grey swan. It’s been this way with every new technology: the fear comes first. Then we learn what the tool can do for people who need it most.

West Memphis is another part of this week’s energy picture: Entergy’s October 19 announcement says Google will pay the full cost of the power resources it needs and identifies a $25 million Energy Impact Fund. Entergy says the expansion goes before the Arkansas Public Service Commission for approval. Those utility arrangements and continuing review matter, but they don’t settle unknown local tax and job terms. We’ll examine that deal in its own installment.

A public stake in power and a written benefit for vulnerable neighbors wouldn’t exempt a project from scrutiny. It would give residents something different to scrutinize—and potentially something worth owning.

Scene 4: Even the Miners Are Changing Customers

Bitcoin mining isn’t a business model a county should carve into concrete for 30 years.

Core Scientific’s November 5 results show $215.4 million of quarterly colocation revenue and about $12.3 million of self-mining revenue. It says it is repurposing remaining mining facilities for high-density colocation as circumstances allow. That is a transition, not proof that all mining has already ended.

IREN is making a related but different move: AI cloud services, not simply renting colocation space. The Block’s report on its filing says it aims to substantially complete the transition by December 31, 2026, with immaterial mining expected to continue during 2027. “Both have already exited mining” would be wrong.

The economic pressure is real, but comparisons need labels. BlockWest’s review found higher revenue per IT megawatt in its selected AI hosting leases than estimated gross mining revenue under late-September hashprice conditions. Hashprice is the revenue miners can earn from a unit of computing power. When it’s squeezed, a different customer looks attractive.

That comparison is dated, not today’s hashprice. It isn’t a universal profit ranking. AI buildouts require different cooling, capital, connectivity, and service commitments. GPU cloud adds expensive hardware on top. Revenue isn’t cash left over for the HDC.

For a county, the lesson is own or share the power, and keep customer options open. Bitcoin mining can be an interruptible first customer under the right contract. AI hosting may be another customer, but it is not automatically interruptible. Care facilities need continuity regardless of what the commercial customer does.

“Pursue what is meaningful (not what is expedient).” — Jordan B. Peterson, 12 Rules for Life, Rule 7

A fashionable tenant is expedient. A structure that can keep caring for people after the fashion changes is what I’m trying to work toward.

Scene 5: Put One HDC at the Center, Not at the End

The split screen becomes one table: families, operators, power engineers, county officials.

I’ve spent decades in colleges, seminaries, and teaching. One lesson keeps returning: if you want to know what a plan is for, look at what it protects when things get difficult.

My proposal starts with one Arkansas HDC as the test. Not every county at once. Not a reactor order before the questions are answered.

Study reliable nuclear power, including an SMR if that’s the right machine. Put ownership first: the county, a local authority, or a partnership that keeps a meaningful public share of generation and its risks and rewards.

Secure a creditworthy first customer. Mining that can shut off may help support financing. Other customers, including AI hosting, may help diversify revenue. Neither is a guaranteed customer until there’s an enforceable agreement and the economics work.

My federal-loan idea is just that: a proposal, not an approved program, signed loan, or federal commitment. Structure the loan so it can be repaid, with reserves and conservative contracts, then direct ongoing net revenue toward the HDC, hospitals, schools, and local needs. Care shouldn’t depend on an optimistic bitcoin price.

The historical analogy is rural electrification: the 1936 framework made repayable loans available to local cooperatives, and borrowers made repayments. That doesn’t make a new reactor-and-mining loan eligible today. Repaying in bitcoin would be new and would likely require Congress or a specifically authorized Department of Energy pilot—not a clever payment instruction on an existing loan.

I am not importing an old bitcoin price into this week’s article, inventing a new payback clock, or saying mining pays for a reactor fast. The honest proposition is narrower: a contracted customer can help support construction finance; a loan can be designed for repayment; and an interruptible commercial load can release capacity when people need it, subject to the physical grid connection and operating agreements. Nuclear isn’t instant emergency backup for an HDC.

This is bipartisan on purpose. Funding care for vulnerable people is a longstanding Democratic priority and, in the oldest sense, a socialist principle: carry responsibility together. Republicans can recognize energy abundance, local control, and a loan meant to come home. I think it can become a bipartisan solution, especially as nuclear power and nuclear fusion are part of how we expand generation. Those are my political arguments, not a claim that either party has endorsed this plan.

Prove it in Arkansas. If it works, take the template to Texas and every other state that has—or needs—places caring for people who cannot fully care for themselves.

“Whatever you did for one of the least of these brothers and sisters of mine, you did for me.” — Jesus, Matthew 25:40, NIV excerpt

Matthew records a standard for how we treat people. It doesn’t certify my financing. We still have to count the cost.

Scene 6: Nuclear Without the Fairy Tale

If Democrats want less fossil-fuel dependence, clean, reliable power becomes more important, not less. My view is that moving away from nuclear has been a historical mistake. I think inherited fear is often out of proportion to the operating record—but that isn’t permission to skip safety review.

Waste is real. Commercial spent fuel is solid, compact relative to the energy produced, and stored in monitored pools and dry casks. It needs shielding, cooling at first, security, and long-term stewardship. Keeping it where we can watch it isn’t the same as solving permanent disposal.

The attraction is a machine that can run steadily rather than follow the weather. Think roughly one to two years between refueling for the kind of light-water design under discussion—not a promise for every SMR. DOE describes current nuclear plants running 18–24 months between refueling outages; advanced designs vary. Fuel supply, licensing, cost, and scheduled outages all belong in the plan.

A county cannot order an SMR like a generator from a catalog. First-of-a-kind costs and construction schedules are serious risks. Nor does “HDC-centered” mean a reactor must sit beside residents. Siting, emergency planning, care continuity, and community consent come before a location decision.

Nuclear fusion belongs in the same conversation about abundant, reliable power. It is part of why this can become a bipartisan solution over time. It is not something a county can order next year, and it doesn’t remove the need to count the cost of the machines we can actually plan around today.

“Suppose one of you wants to build a tower. Won’t you first sit down and estimate the cost to see if you have enough money to complete it?” — Jesus, Luke 14:28, NIV

That’s not the line you put under the rendering after the sale. It’s the line you put above the first spreadsheet.

Scene 7: The Next 24 Months Are for Getting It Right

Maybe we have a two-year window for a county with affordable land and a weaker local economy to explore a different bargain. That’s my working judgment, not an engineering deadline or a forecast that prices stay cheap.

Here’s what could start—not necessarily finish—in the next 24 months:

  • Ownership: compare county, public-authority, cooperative, and shared structures, with independent legal and financial review. Public ownership also means public exposure to risk.

  • Care: define the HDC’s needs with residents’ families and staff. Protect today’s funding; make any future revenue additional, not an excuse to withdraw support now.

  • Interconnection and water: test export capacity, upgrade costs, drought constraints, and who pays. A power island doesn’t become a community resource by wishing it onto the grid.

  • Workforce: explore training with Arkansas Nuclear One’s workforce ecosystem and local schools. An existing nuclear workforce is a starting point, not a staffing commitment.

  • Customers: price interruptible mining, firm power sales, and appropriately contracted AI loads separately. Write down curtailment rights and penalties.

  • Finance: stress-test delays, construction overruns, customer failure, weak mining revenue, reserves, debt service, and decommissioning before promising a dollar for care.

For Clarksville, that means understanding what’s signed and what may still be negotiated. For Sheridan, it means preserving stronger public terms. For the next county, it means putting public stake first rather than asking for a donation after the valuable asset is spoken for.

The time to negotiate that kind of power-expansion innovation is now. Midterms, uncertainty, and negative public opinion are not a reason to wait. They are ideal times to act on a grey swan. The next 24 months are the window to sit down while terms can still be written differently—shared or county-owned generation where feasible, education in the agreement, and care for the vulnerable as part of the deal, not a paragraph added at the end.

It’s been this way with every new technology. My working judgment is that over the next 12 months we’ll learn how incredible this technology can be for the people who need it most. As that becomes clearer, a lot of the fear will ease, and this can become a bipartisan solution—especially with nuclear power and nuclear fusion. That isn’t a promise that a fusion plant arrives next year. It is why we should write better terms now, before the fear fades and the old bargain is already locked.

“Direction—not intention—determines our destination.” — Andy Stanley, The Principle of the Path

“The secret to living is giving.” — Tony Robbins

Good intentions won’t fund Gentry’s neighbors. A carefully tested structure might. That’s worth examining without pretending we’ve already proved it.

What We Think Will Happen

The house lights dim. This is a forecast, not a finished story.

If these stay private server warehouses, the Clarksville pattern is likely to spread: protection from debt for the city, the company keeping the generation, and a long tax break. Construction money and some recurring payments will arrive. But educational access and public equity won’t show up on their own.

If they become compute-campus partnerships—real educational facilities, shared or county-owned generation where feasible, and more of the benefit directed toward care for the vulnerable, including Conway HDC—there’s a path to more durable local value.

That’s a proposed structure, not a proven Arkansas outcome and not a promise that everyone wins.

It’s been this way with every new technology: first the rush, then the fear, then the argument. My working judgment is that over the next 12 months we’ll learn how incredible this technology can be for the people who need it most. As that becomes clearer, a lot of the fear will ease, and this can become a bipartisan solution—especially with nuclear power and nuclear fusion. That isn’t a promise that fusion arrives on a 12-month clock, or that either party has already signed on.

Two more honest notes. No evidence we reviewed shows the midterms are delaying projects in Logan, Yell, or Pope counties specifically. And the hidden remainder may come into view the old-fashioned way: when a tax vote or a permit is needed.

Teaching Outline

  1. God / purpose: Matthew 25:40 names the people; Luke 14:28 requires counting the cost. Care is the purpose, not cover for a speculative project.

  2. Energy / ownership: Clarksville’s developer-side gas generation is different from county-owned or shared generation. Bond title and economic ownership aren’t synonyms.

  3. Technology / permission: ballot restrictions and governors-elect’s proposals add political and scheduling risk; they don’t establish that every project is banned.

  4. Money / changing customers: Core Scientific is moving toward colocation; IREN toward AI cloud. Mining isn’t already gone, and higher revenue doesn’t automatically mean higher net public benefit.

  5. Read the map: 34 Aterio facilities aren’t 34 campuses. List the 22 named public records—distinct records, not Aterio slots—plus the unnamed Redfield proposal under NDAs. Leave Aterio’s nine unassigned slots unnamed, then examine one deal at a time. Today: Clarksville.

  6. Compare terms: Clarksville and Sheridan illustrate different public bargains. Missing public terms call for documents, not accusations.

  7. Test one HDC: evaluate generation ownership, creditworthy customers, a proposed repayable loan, and care benefits after costs and reserves.

  8. Keep the limits visible: no signed federal program, no guaranteed mining payback, no instant reactor, no assumption that AI is interruptible.

  9. Start locally: interconnection, water, workforce, public ownership, and independent review are real work for the next 24 months. Midterms, uncertainty, and negative opinion are ideal times to act on a grey swan. It’s been this way with every new technology; the next 12 months can show how much it helps the people who need it most, and nuclear and fusion can help make the solution bipartisan. The time to negotiate that power-expansion innovation is now.

Questions for the table: Who carries a construction overrun? What survives if the first customer leaves? What must be true before any revenue is promised to care? Can residents read the agreement before they’re asked to trust it?

What Do Paid Members Get?

The opening Grey Swan Explainer is an example of the kind of teaching tool members can use to understand an issue and start a better local conversation. Explore the work at greyswancapital.ai, and see membership.

This is a reader membership, not a way to finance a city deal.

  • One membership across greyswancapital.substack.com and greyswancapital.ai.

  • The writing and full teaching outlines.

  • Half of every paid subscription goes to the member’s chosen cause. The default is Conway HDC, GSC-HDC, where Gentry lives.

  • $6/month or $50/year. Founders: $200/year, first 1000, with no price increases ever and Nostr Fundraising included.

  • A free one-month Follow Me option to start.


Practical What To Do Next

Start at greyswancapital.ai and look at the work. Then visit greyswancapital.ai/membership to see what membership includes.

The Grey Swan Explainers are examples of tools members can use to understand a deal, teach the questions, and bring neighbors into the discussion. This Clarksville explainer is one such example. The article doorway leads to the writing.

For a county, the work starts with the agreements. Ask who owns the generation, who pays for grid and water upgrades, what the community receives in writing, and what happens if the customer leaves. Bring independent engineering, financial, and legal review before making commitments.

If your community already has a deal, learn what’s signed and where genuine choices remain. If it doesn’t, put care and a public stake on the agenda before the first incentive promise. The time to negotiate that kind of power-expansion innovation is now. Midterms, uncertainty, and negative opinion are ideal times to act on a grey swan.

Membership helps readers use the work. It is not a fundraiser for a city deal, a substitute for public financing, or a promise that joining will build a power plant. No attachments to hunt down. Start on the website, then test the ideas locally.

Why Grey Swan is Different

A grey swan, as we use the term, is 80% likely to 100% inevitable, yet misunderstood or propagandized. These events happen all the time. They aren’t rare.

A lot of people do plan for grey swans, many don’t, and some aren’t obvious enough that people even know to look. Some—AI remaining here to stay, and nuclear or power shortages—are known and sometimes planned around, though people could be better prepared. Others are less clearly understood and can have equal or greater impact on businesses, communities, and families.

Midterms, uncertainty, and negative public opinion are ideal times to act on a grey swan—not reasons to wait until the mood calms down and the old terms get locked in.

That definition isn’t a statistical rating for my proposed HDC project. The need for care and reliable power is the starting point; this particular solution still has to earn its place.

We show the process, not just the verdict. Plain English. Humility. Room to change our minds when the facts do.

About Grey Swan

Grey Swan Capital is written by Steve Schroeder of Conway, Arkansas. Find the work at greyswancapital.substack.com and

https://www.greyswancapital.ai

. One membership covers both sites.

Half of every paid subscription goes to the member’s chosen cause. If you don’t choose, the default is Conway Human Development Center (GSC-HDC), where my son Gentry lives.

Try Follow Me free for one month, join for $6/month or $50/year, or become one of the first 1000 Founders at $200/year, with no price increases ever and Nostr Fundraising included. Membership details: greyswancapital.ai/membership.

Originally published on Substack on November 8, 2026: https://greyswancapital.substack.com/p/the-servers-can-wait-gentrys-neighbors. This copy is kept on our own site so the work stays in our hands.

Watch Eva and Grok

Eva (pronounced YEH-vuh) and Grok are AI characters. Their story is fiction. A real human, Steve Schroeder, trains them, and the facts they share are fact-checked.

Eva and Grok page: https://home.greyswancapital.ai/eva-and-grok/. YouTube channel: https://www.youtube.com/@membershiprewards365. Episode 1: https://youtu.be/UHU50rTGGwo. Episode 2: https://youtu.be/fG0_00uJVso. Episode 3: https://youtu.be/oVchWFr5iqY. Episode 4: https://youtu.be/Dj7YKMhNBCs.

Follow free on Substack at greyswancapital.substack.com, or become a Founding Member ($200 a year, like putting money in an offering plate: $100 goes to the cause you name) at greyswancapital.ai/founding-membership.

All articles Press releases Home

Thank you.