Skip to content
An open protocol for community-owned care · Boulder is proving it

The infrastructure aging forgot.

An open protocol and a member-owned federation, so any community can run — and own — its own care.

63 million Americans are family caregivers (AARP and National Alliance for Caregiving, Caregiving in the US 2025) — no badge, no record in the EHR, no measurement of outcomes they produce. Home care agencies lose about 75% of their caregivers every year (Activated Insights, 2025 Benchmarking Report). The gig economy is Standard Oil trying to deliver electricity one household at a time.

A century ago, a cooperative grid fixed the same infrastructure gap in a generation. Vimty applies that model to aging care: the shared protocol, and a federation the communities themselves own. Boulder is Node One — being built now. Today co-op.care takes sign-ups, runs a family assessment, and keeps a time bank; paid care begins once it is licensed and staffed, and caregivers become worker-owners when they are paid. What it is not yet is a national grid. This is the standard and the invitation; the license opens when Boulder is proven.

~900
electric cooperatives operating in the US today (NRECA, 2026) — proof the model lasts
42M
people served by consumer-owned electric cooperatives, across 56% of the American landscape (NRECA)
1936
Rural Electrification Act — about 1 in 10 farms had power in 1930; more than 9 in 10 by 1953. Same model. Same form.
Read the model Build the next node — when it's ready Partner with Boulder — Node One

Here for a parent, not a protocol? The family door onto this same grid is co-op.care — see what your family qualifies for, and join the Boulder circle.

Vimty in 15 seconds

Text version:
  1. Aging care runs on people nobody counts. Family and paid caregivers hold it together.
  2. In 1930, 1 in 10 farms had power. By 1953, 9 in 10. Co-ops wired what the market would not.
  3. The same model, for care. Caregivers as worker-owners of their own cooperative.
  4. Boulder is Node One. It is not yet a national grid. It is an invitation.
  5. Vimty. An open protocol for community-owned care. vimty.com

Caring for someone now? Start at co-op.care. Want a node in your community? Put your city on the map.

The proof of concept

A century ago, the same gap. A cooperative grid solved it.

1930
~10%
of US farms had electricity
1936
REA
Rural Electrification Act signed
1953
90%+
of US farms had electricity
Today
~900
electric co-ops still running
2026
Care grid
Boulder — Node One forming

In 1930, only about one in ten American farms had electricity (Richmond Fed, Econ Focus, 2020). Private utilities ran lines to profitable urban density and stopped. Rural communities were simply not worth the capital per mile of wire. The market verdict: you do not get power.

Franklin Roosevelt signed the Rural Electrification Act in 1936. The mechanism was not a government utility. It was government-backed cooperative loans to communities that would form their own cooperatives, build their own lines, and own the infrastructure they depended on.

By 1953, more than 90% of American farms had electricity (NRECA). The Rural Electrification Act turned 90 in 2026. It has outlasted every gig-economy power scheme because its incentives compound: the member-owners who need the electricity are the same people who own the grid.

The cooperative grid — then and now

~10%
US farms with electricity in 1930 (Richmond Fed). The infrastructure gap before the cooperative grid.
90%+
Farm electrification by 1953 (NRECA). Under a generation.
~900
Electric cooperatives operating today, serving 42 million people across 56% of the American landscape (NRECA fact sheet, 2026).
90yr
Since the Rural Electrification Act was signed on May 20, 1936. The cooperatives it financed are still running.
The gig economy is Standard Oil trying to deliver electricity one household at a time. The cooperative grid is the Rural Electrification Act. We know which one is still running 90 years later. — the Vimty thesis

The care grid — same model, same form

Generation

Caregivers produce care hours. Worker-owned cooperatives employ them W-2. The same shift that delivers care produces the clinical observation.

co-op.care →

Transmission

Planned: shared care software, physician review of AI output (HarnessHealth), and a user-owned memory (chanio) — the protocol layer, to be licensed at cost to every node. The wires that carry it.

harnesshealth.ai →

Distribution

Local cooperatives — the last mile into each home. Community-owned, locally governed. Each node owns its own equity. The grid interconnects them.

Start a node →

Metering

Care-hour credits — the kilowatt-hours of the care grid. Earn a care hour in Boulder. Redeem it in Amsterdam. The ledger keeps the account.

How care credits work →

Standards

Omaha System to FHIR mapping — the voltage standard. Designed so an observation from any node can travel to any clinician.

HarnessHealth →

Ownership

Member-owners hold equity in their local cooperative, not in a holding company above them. Democratic governance. The incentive structure that made the REA last.

How ownership works →
Seven layers · what the network is designed to produce

A coherent organization of layers to humanity.

This is the design, not a running network: agents across seven layers of a member's life, with a physician behind anything clinical. The care comes first. As the network matures, that care would also leave a record, signed by a physician only where a physician actually reviewed it, and used for anything beyond care only with member consent. It is a product of the care, never the reason for it.

Layer 1 · Arrival
Members enter the network
Designed: intake from the condition sites plus cooperative onboarding.
Layer 2 · Belonging
One member identity
Designed: one identity, a cooperative member share, and a governance vote, so every record points to the same person. ComfortCard shows a demo wallet card today.
Layer 3 · Daily flow
Sage runs in the background
Designed: persistent context that any AI model can read, so each interaction adds to the record without more forms.
Layer 4 · Meaning
Family timeline + values clarification
Designed: advance directives, care goals, and the family story. The scaffolding that makes Layer 6 humane instead of transactional.
Layer 5 · Body your feed
A record the care leaves behind
Designed: each observation time-stamped, fingerprinted on hashcare, and member-consented, and signed by a physician only when a physician has reviewed it. None exist yet; there is no paid care to observe. Any use beyond care waits for member consent.
Layer 6 · Care production engine
Worker-owned cooperative caregivers
Planned: the same caregiver shift that delivers the care also leaves the record. Worker-owned, so the value stays with the people who create it. Paid care in Boulder begins once it is licensed and staffed.
Layer 7 · Mind
Compounding personal context
Designed: the member's own AI context, owned by them. chanio runs this today on the founder's own Mac, not yet for members.

Brands are doors; layers are the rooms. Layer 6 operations — the care — sustain everything; Layers 1–4 + 7 hold the member; Layer 5 is the record the care leaves behind. That is the design. Today Boulder is the first node being built, and nothing is called attested unless a physician signed it.

Join the cohort discussion How ownership works

The unmeasured layer

Aging is held together by people the system does not count.

The family caregiver is the single most important variable in whether a discharge holds, a medication is taken, a fall is prevented, a hospital readmission is avoided. None of that caregiving shows up on a claim, a chart, or a board-level dashboard.

It does show up — in strain, in absenteeism, in cost shifted to emergency departments, and in outcomes the health system takes credit or blame for without knowing the real cause. ARCHANGELS calls caregiving intensity the vital sign organizations did not know they were missing. That framing is correct. The implication is that the industry is running blind on the variable that matters most.

When we talk about aging infrastructure, we do not mean hospitals, senior-living towers, or another app. We mean the human layer that is already doing the work. It needs a record, a structure, and a way to compound.

Why current fixes fail

The two dominant answers — the gig economy, and the chatbot — are each structurally broken.

1. The gig model cannot retain the caregiver.

U.S. home care runs on about 75% annual caregiver turnover (Activated Insights, 2025 Benchmarking Report). Matching a family to a worker who will churn inside the year is not infrastructure. It is a fee on catastrophe.

Platforms are applying AI to scheduling to hold caregivers longer. That helps. But the caregiver still builds value inside somebody else’s platform, and the improvement is priced in to the platform, not the worker.

2. The unsupervised chatbot cannot be trusted with medicine.

A 2026 wave of peer-reviewed studies is converging on the same conclusion, four studies in three journals in the space of months: chatbots cannot be the patient-facing clinical interface without a physician in the loop.

BMJ Open (2026)
Generative AI chatbots and medical misinformation — accuracy, referencing, readability audit
Five popular chatbots audited on health questions. Nearly half (49.6%) of responses were problematic, about one in five highly problematic. Citations were frequently incomplete or fabricated.
JAMA Network Open (2026)
Large Language Model Performance on Clinical Reasoning Tasks
21 frontier models across 29 clinical questions. Conclusion: current LLMs remain limited in early diagnostic reasoning and cannot yet be relied on for unsupervised patient-facing clinical decision-making.
Nature Medicine (2026)
Reliability of LLMs as medical assistants for the general public
Randomized, preregistered study. LLMs in lay-public hands identified the relevant condition in fewer than 34.5% of cases — no better than a control group. Patients could not guide the model to the right questions.
Nature Medicine (2026)
ChatGPT Health performance in a structured test of triage recommendations
Among gold-standard emergencies, the system undertriaged 52% of cases, with inconsistent activation of crisis safeguards. Safety concerns warrant prospective validation before consumer-scale deployment.
I warned three years ago that these systems were “purveyors of authoritative bullshit” that should not be trusted. That is still true — and it very much applies in medicine. — Gary Marcus, April 2026, revisiting a 2023 warning in light of four new peer-reviewed studies

The gig economy churns the worker. The unsupervised chatbot hallucinates the plan. Neither has the structural shape the aging population needs.

The shift under all of this

Services are the new software.

Julien Bek of Sequoia framed the shift cleanly in March 2026. The next wave of AI winners will sell outcomes, not tools. Pure software gets replicated by incumbents, underpriced, or built in-house. The durable companies are full-stack — they own the care model, the patient relationship, the clinical data, and the operations. AI embedded across that stack reinforces the service, instead of being the service.

A copilot sells the tool. An autopilot sells the work. — Julien Bek, Sequoia Capital, “Services: The New Software” (March 5, 2026)

In healthcare the shift is already moving: services companies are embedding AI across a care-delivery stack they own end-to-end. The bet is that the margin profile of a services business changes when AI takes real cost out of care delivery.

Vimty’s argument is one step past Sequoia’s: if services are the new software, then the ownership structure of the service is the moat. Capital-owned services compound for the fund. Worker-owned services compound for the worker, the family, and the community around them.

Community leverage = the moat

Make the worker an owner. Make the community the customer of last resort. Then embed the AI.

The home-care-specific proof already exists. Cooperative Home Care Associates (CHCA) in the Bronx has been running the model Vimty is arguing for since 1985 — more than 1,600 caregivers and staff, about half of them worker-owners (CHCA), with turnover reported far below the industry’s. Forty years. One of the largest worker cooperatives in the United States.

Industry baseline
~75%
Annual caregiver turnover, home care agencies (Activated Insights, 2025 Benchmarking Report)
Cooperative model
~24–38%
Annual turnover reported for worker-owned home care cooperatives (PHI, 2024; The Nation, Aug 2026)

The Mondragon Corporation is the broader proof the worker-owned model scales. Founded in the Basque Country in 1956, it reported €11.06B in 2023 sales and about 70,500 people employed (Mondragon Annual Report 2023). Co-operatives UK reports that about 80% of co-operatives survive their first five years, vs. 44% of other businesses (The Co-operative Economy, 2018). Worker ownership is not idealism. It is a survival characteristic.

1,600+
CHCA caregivers and staff in New York, about half of them worker-owners (CHCA). Forty years in home care specifically.
~24–38%
Annual turnover reported at worker-owned home care co-ops, against about 75% across the industry (PHI, 2024; The Nation, Aug 2026; Activated Insights, 2025).
80% / 44%
Five-year survival, co-operatives vs. other businesses (Co-operatives UK, 2018).
§1042
U.S. tax-code mechanism letting a founder sell to a worker-owned cooperative and defer capital gains. A quiet, legal exit path.

What this looks like in aging care.

The gig stack
Caregiver is a contractor. Data is the platform’s. Equity is the fund’s.
  • ~75% annual caregiver turnover (industry baseline)
  • Family rebuilds the relationship every year
  • Care data sits in a vendor portal and evaporates on churn
  • Upside accrues to the platform and capital stack
  • AI is added to keep contractors from leaving
The cooperative stack
Caregiver is an owner. Data is community-held. Upside is shared.
  • W-2 employment + equity in the cooperative
  • The relationship compounds — same neighbor, year after year
  • Clinical-grade record owned by the family, portable across any clinician
  • Profit reinvests into care capacity and caregiver compensation
  • AI is embedded to make the owner more valuable, not the contractor cheaper

A platform uses AI to hold the worker. A cooperative makes them the owner. The moat is not the model — it’s the legal form.

The replication proof

Buurtzorg proved a shared protocol scales. Without a franchise.

In 2006, Jos de Blok started Buurtzorg, a Dutch nonprofit, with one small team of nurses. No venture capital. No franchise agreement. A back office of a few dozen people. One shared protocol: self-managing teams of up to 12 nurses and common technology, with every team in control of its own work.

By 2022: more than 10,000 nurses in about 900 self-managing teams across the Netherlands, and the model now in use in 25 countries, including the US, UK, Japan, Sweden, and Australia (Buurtzorg International). A 2009 Ernst & Young study found Buurtzorg met patients’ needs using about 40% of the authorized care hours, against about 70% for other agencies (Commonwealth Fund, 2015).

The replication mechanism is not a brand extension. It is a shared protocol. Buurtzorg maintains the common technology layer; each team runs its own work. Vimty’s step past Buurtzorg is ownership: each community would own its cooperative — the employer relationship, the family relationships, and the equity.

2006
Founded in the Netherlands by Jos de Blok as a nonprofit. One small team. No venture capital.
10,000+
Nurses in about 900 self-managing teams across the Netherlands alone (2022).
25
Countries where the Buurtzorg model is in use (Buurtzorg International).
40% vs 70%
Share of authorized care hours used, Buurtzorg vs. other agencies (Ernst & Young, 2009).

Vimty’s architecture is designed for the same replication. co-op.care is building the first node, in Boulder. The stack is the protocol. The next community adopts it, builds its own cooperative, and the network grows — not as a franchise, but as a federation. Each node locally owned. Each node on the same technology and physician-review layer.

Boulder, Colorado
Limited Cooperative Association — Colorado Title 7
Node One — building now
Netherlands
Coöperatie U.A.
Form mapped
Canada — Quebec
Coopérative de travail (EVA homecare precedent)
Form mapped
United Kingdom
Co-operative Society — FCA registered
Form mapped
Japan
労働者協同組合 — Worker Cooperative Corporation Act (2022)
Form mapped
Australia
Co-operatives National Law
Form mapped
Your city
Put your city on the map
Start a node →

When a community is ready to join

One agreement. Three obligations. No equity extraction from the community — ever.

Nothing here is for sale today. These are draft terms for once there is a proven network to join; we publish the shape now so the model is transparent, and any amounts will be set with the first communities. Every community that federates signs a Protocol License Agreement. It is not a franchise agreement: the community owns its own cooperative, its own employer relationships, and its own equity. The PLA covers the three things that keep the network coherent: access to the shared technology stack, a care-hour solidarity contribution that funds the next node, and a one-time setup that covers onboarding and legal scaffolding.

Obligation 1
Share
Solidarity contribution
A small share of each node’s member revenue would flow to the network solidarity fund. The fund seeds the next community’s cooperative formation, covers legal scaffolding for new nodes, and maintains the shared technology stack. The mechanism is the same one the rural electric cooperatives used to build out lines into communities that couldn’t yet pay the full capital cost.
Obligation 2
Setup
One-time node setup
Would cover onboarding to the shared software, cooperative legal scaffolding in your jurisdiction, the initial Omaha-to-FHIR mapping, and onboarding to physician review. No recurring platform fee on top. The solidarity contribution is the ongoing network cost — no SaaS invoice.
Obligation 3
Open
Credit portability — the network promise
Every federated community agrees to honor care-hour credits earned at any other node in the network. A member who earns hours in Boulder can redeem them in your city. The care credit is as good as a kilowatt-hour was in the REA grid: a unit of the network’s collective capacity, redeemable anywhere on the wire.
What the PLA does not require
No equity transfer to Vimty Your cooperative’s equity belongs to your worker-owners. Vimty takes no stake in your community’s cooperative. The solidarity contribution flows to a fund, not to a central holding company.
No per-member data licensing fee No per-member data fee, no per-observation charge, and no API usage billing. The solidarity fund covers shared infrastructure.
No central governance override Your cooperative is governed by your members, under your jurisdiction’s cooperative law. The PLA sets minimum protocol standards (Omaha-to-FHIR, attestation, credit portability). It does not override your governance.
No exclusivity restriction A federated node can operate additional services, community programs, or partnerships outside the Vimty protocol. The PLA covers the shared stack and credit portability. Your cooperative’s scope is yours.
The equity is always local. The protocol is always shared. The solidarity contribution is the wire maintenance that keeps the grid connected. — Protocol License Agreement, draft Vimty federation terms
Put your city on the map How co-op.care ownership works

The build — three layers, at different stages

One operator. One physician-review layer. One user-owned memory. None finished yet.

Vimty is not a product. It is the argument for a three-layer stack, each layer a separate build, each answering a different structural failure. Together they remove the architectural harms (gig labor, unattested AI output, vendor-held records) before attempting to add capability on top.

Layer 1 — The operator
co-op.care
co-op.care
A worker-owned cooperative for aging care, forming in Boulder. Caregivers will be W-2 worker-owners once paid care begins. Today: sign-ups, a family assessment, and a time bank. Paid care begins once it is licensed and staffed.
Visit co-op.care
Layer 2 — Physician review
HarnessHealth
harnesshealth.ai
Designed so no AI draft reaches a patient until a named physician signs it or declines. The answer to the 2026 chatbot studies: AI drafts, a physician decides. Today: a sandbox that runs on synthetic cases; production reviews open when the queue goes live.
Visit harnesshealth.ai
Layer 3 — The memory
chanio
chanio.com
The user-owned memory every AI in the stack reads from, and none of them keep. Your notes, chats, and decisions as plain files on your own device, readable by any model. Any AI can read it. No AI can hold it. Today: it runs nightly on the founder’s Mac; phones come next.
Visit chanio.com

Operator + physician review + memory. The three layers are separately buildable and structurally interdependent. No one of them works without the other two.

The pre-institutional moment

Aging care today is where public health was in 1890.

Enough scientific foundation to be serious. Not enough institutional infrastructure to be systematic. A credibility spectrum that runs from rigorous research to influencer advice. No agreed scoreboard for what aging well even looks like.

Public health’s early wins came from removing harms first — clean water, vaccines, sanitation — before it tried to add capability. The architectural harms in aging today are fragmented data, gig labor, and unattested AI clinical output. Vimty’s thesis is that those harms have to be removed first. The capability layer comes next, and safely.

MIT AgeLab launched a Longevity Preparedness Index in October 2025 to try to measure what preparedness for a longer life looks like at the individual level. Joseph Coughlin’s broader argument in The Longevity Economy is that an aging society is the most misunderstood market of our era. Vimty’s contention is that individual preparedness is not enough. The unit of preparedness is the community, and the infrastructure is cooperative.

Community leverage is how removal happens. Capital alone cannot retain the caregiver, audit the chatbot, or hold the record. A cooperative can.

The real token

Earn care now. Redeem it later. Anywhere in the network.

In 1995, Japan’s Sawayaka Welfare Foundation created Fureai Kippu — loosely, “caring relationship ticket.” The mechanism: help an elder today, earn a time credit, keep it in a mutual-aid ledger, redeem it when you or your family need care — or transfer it to a parent in another city through a clearinghouse.

It is one of the longest-running care tokens there is. Not speculative. Not blockchain-dependent. A cooperative-ledger time credit backed by actual care capacity in every affiliated organization. One hour given earns one hour redeemable.

Contribute ten hours. Get the future you want. — The Vimty network at scale: mutual aid as infrastructure

Vimty extends this model to a federated cooperative network. co-op.care’s time bank is live: an hour given is an hour banked. Credits are member-owned, cooperative-held, and — as the network federates — redeemable at any affiliated community. When you age, wherever you are, the hours are yours.

Japan, since 1995
Fureai Kippu — the original cooperative care credit
Time-banking for elder care. Help a neighbor; earn a credit. Clearinghouses move credits from one side of Japan to the other. About thirty years of evidence that care-hour portability can be run administratively. (Sawayaka Welfare Foundation)
Colorado, 2026 — being built
co-op.care — the first node
Boulder is building the model: W-2 worker-owners once paid care begins, cooperative equity, and a time bank that is live today. The first node is meant to prove the mechanism. Communities that follow would join the same credit portability.
The settlement layer
One hour in. One hour out. Anywhere in the network.
A member who earns care credits in Boulder can redeem them in any federated community once that community is live. The value of the network compounds with every new node — mutual-aid network effects, without the speculation of a digital token.
The governance
Cooperative ledger — not blockchain
Time credits are tracked in the cooperative’s own member ledger, member-verified and governed by each cooperative. Designed to be auditable and portable. Whether credits can move across borders is a legal question each jurisdiction still has to answer.
Where the equity lives — and where it does not

The equity stake in this network is not central. A cooperative owned by a central entity is not a cooperative. Vimty is the protocol layer — shared care software, HarnessHealth physician review, chanio memory — designed to be licensed to each community at cost. No equity in the infrastructure layer. No central extraction.

Each community that federates forms its own locally governed cooperative entity. The stake — the real token — is in that entity. Its workers hold equity in their community’s cooperative, not in a holding company above them.

Community Cooperative entity Status
Boulder, Colorado Limited Cooperative Association (LCA) — Colorado Title 7 Being built
UK Co-operative Society (registered with FCA) Form mapped
Germany Eingetragene Genossenschaft (eG) Form mapped
Netherlands Coöperatie (U.A.) Form mapped
Japan 労働者協同組合 — Worker Cooperative Corporation Act (2022) Form mapped
Australia State-registered Cooperative (Co-operatives National Law) Form mapped
Canada Coopérative de travail — Quebec (strongest ecosystem; EVA co-op is a live homecare precedent) Form mapped
Your community Your local cooperative form — we help you find it Express interest →

Boulder is the first node, still being built. Each subsequent city that joins operates its own cooperative, governed by its own members. The Vimty stack is the shared infrastructure. The equity is always local.

Notes: Germany eG requires membership in a cooperative audit association (Prüfungsverband), adding formation overhead. Japan’s Worker Cooperative Corporation is new (2022) and carries minimal homecare operational track record so far. Canada’s strongest cooperative ecosystem is Quebec.

How co-op.care ownership works

The real token is a care hour, or a stake in your community’s cooperative. Locally held. Locally governed. Not pooled at the center.

Where this comes from

A fifteen-year attempt to get people to face aging — and the reason it kept failing.

This is not a new idea. In 2011 it was Vimty — an all-digital advance directive, built to do one stubborn thing: get people to talk about the end of their lives before a crisis decided for them. To make it real, it had to solve digital notarization across state lines — assembled around interstate commerce, before most of the country allowed remote online notarization at all. The technology worked. The problem did not move.

In 2016 it became CareGoals — the same mission, aimed at the reimbursement rails this time: the Advance Care Planning codes and the Annual Wellness Visit, the two places Medicare will actually pay a clinician to have the conversation. The codes existed. The conversation still didn’t happen at scale. Many years and several good tools later, it is still not solved.

The lesson took a decade and a half, and it is the whole reason for Vimty now: the problem was never the document or the billing code. It was that no one owned the relationship or the record. A form can’t make a family talk. A code can’t retain the caregiver who’d be in the room. Only a community that owns its own care — the people, the record, and the outcome — can hold a conversation that outlasts a single visit.

Vimty is the fifteen-year mistake finally worth making — because for the first time the pieces exist: AI to clear the friction, and a cooperative to own what’s left.

The grid is open. Plug in.

The Rural Electrification Act built a cooperative grid now serving 42 million people across 56% of the American landscape (NRECA). It is 90 years old and still running. Boulder is node one of the care grid, being built now. Two doors: a family looking for care starts at co-op.care; a community that wants its own node puts its city on the map.

Build the next node

The replication protocol is the same one that built the rural electric grid: a shared technology layer, a cooperative legal form, and locally owned infrastructure. Put your city on the map now; the stack and legal scaffolding open once Boulder is proven. Boulder is node one. Your city could be node two.

Start a node in your city See the Boulder model
Legal forms mapped so far: UK, Germany, Netherlands, Japan, Australia, Canada (Quebec). Each has a cooperative form. The technology would be the same. The cooperative is locally owned everywhere.