Open Cooperative Infrastructure for the AI Era

What if the help you need
is already nearby?

Every day, millions of cooperative connections fail — not because people are unwilling, but because neither party knows the other exists. The retired accountant three blocks from the struggling freelancer. The idle kitchen next to the neighborhood with no gathering space. The bilingual nurse whose translation skills could help dozens of families — if anyone knew to ask.

This isn't a technology problem. It's a perception problem — and it's costing humanity trillions.

Try the Interactive Demo Read the Specification

$1.3T

Estimated Gap

42%

Hidden Expertise

83%

Jobs Via Weak Ties

Protocol Layers

Experience It

Try the demo to see matching in action

→

Understand It

Read the specs and evidence base

→

Challenge It

Vote on open questions, debate positions

→

Build It

Join a working group, write code or specs

The Invisible Crisis

The Latent Value Gap

A retired engineer who could diagnose your drainage problem in ten minutes lives three miles away. A college student builds websites for fun while the food bank next door drowns in spreadsheets. Neither knows the other exists. No marketplace, social network, or job board can find these matches — because neither party would think to search.

👩‍🔧

Maria, 62

Retired mechanical engineer

Volunteers at the library. Tends her garden. Feels a bit purposeless since retirement.

✦ 35 years of drainage & HVAC expertise she doesn't think of as "skills"

👨‍👧

James, 34

Single dad, warehouse worker

Water pooling near his foundation. Contractor quotes: $3,000. He figured he'd just deal with it.

✦ Skilled woodworker — Maria's library needs bookshelves

This connection died in silence.

Maria could save James thousands. James could build what Maria's library needs. Three miles apart. Invisible to each other. This happens millions of times, every day, everywhere.

The Deeper Question

Why can't we see each other?

The latent value gap isn't a bug in our platforms. It's a consequence of how human cognition, social structure, economic systems, and institutional design interact to make cooperation invisible. Understanding why the gap exists is essential to designing something that actually closes it.

Psychology

Psychologists call it the curse of knowledge: once you've internalized an expertise, you literally cannot remember what it was like not to have it. The retired accountant doesn't think of tax optimization as a skill because to her it's just... obvious. Simultaneously, the struggling freelancer can't articulate what he needs because he doesn't know what he doesn't know — a phenomenon called meta-ignorance.

These aren't character flaws. They're universal cognitive limitations. Dunning-Kruger works in reverse here: genuine experts systematically undervalue their rarest capabilities. The person most able to help is the person least likely to offer, and the person most in need is the person least likely to ask.

The skills that feel most natural to you are precisely the ones that are most invisible to you — and most valuable to someone else.

Kruger & Dunning 1999; Camerer, Loewenstein & Weber 1989; Fischhoff 1975

Sociology

Mark Granovetter's foundational discovery — that weak ties, not strong ones, are the primary conduits for new opportunities — is now one of the most replicated findings in social science. Raj Chetty's team confirmed it at planetary scale: analyzing 21 billion Facebook friendships, they found that cross-class connection is the single strongest predictor of upward economic mobility — stronger than school quality, family structure, or local job markets.

But here's the cruel irony: every social platform we've built optimizes for homophily — connecting you with people like you. Recommendation algorithms maximize engagement by reinforcing existing patterns. The connections with the highest cooperative potential — the cross-boundary, complementary, unexpected ones — are systematically filtered out by the systems designed to "connect" us.

The most valuable connections in your community are the ones that would never appear in your feed, your search results, or your social graph. They exist in the space between networks — and nothing is designed to find them.

Granovetter 1973; Rajkumar et al., Science 2022; Chetty et al., Nature 2022; McPherson et al. 2001

Economics

Market economies are extraordinary engines for matching supply with demand when both sides can be articulated and priced. But latent value defies articulation. You can't post a want-ad for "someone whose life experience maps to a problem I haven't identified yet." You can't price "the feeling of being useful again after retirement." There is no marketplace for complementarities that neither party recognizes.

Economists call these missing markets — situations where mutually beneficial exchanges don't happen because the transaction costs of discovering and arranging them exceed the expected returns. Nobel laureate Elinor Ostrom spent her career proving that communities routinely solve these problems through cooperative governance — but only when they can see each other. The question isn't whether cooperation works. It's whether we can make the invisible visible at scale.

The $1.3 trillion "latent value gap" isn't the value of goods that fail to trade. It's the value of cooperative relationships that fail to form — because markets have no mechanism for discovering them.

Ostrom 1990; Coase 1960; Akerlof 1970; Arrow 1962

Philosophy

We live in the most "connected" era in human history. Average social media users maintain hundreds of digital connections. Yet loneliness, civic disengagement, and institutional distrust are at record levels. Robert Putnam's Bowling Alone thesis has only accelerated: we have more contacts and less community.

The distinction matters. Connection is a link between two nodes. Cooperation is a mutually beneficial exchange that creates value neither party could create alone. Every social platform we've built optimizes for connections — follows, friends, endorsements. None optimizes for the asymmetric, complementary, cross-boundary relationships where cooperative value actually lives. We've been building the wrong thing.

Six billion people carry smartphones, each one a node in a planetary network. But a network of nodes is not a community. A community is a network of capabilities and needs — and that network is almost entirely unmapped.

Putnam 2000; Turkle 2011; Sennett 2012; Arendt 1958

Evolutionary Biology

Humans evolved extraordinary capacities for cooperation — far beyond what any other primate achieves. Direct reciprocity, indirect reciprocity, group selection, and kin selection together created a species that builds civilizations through cooperative behavior. But these mechanisms evolved for groups of 50–150 individuals (Dunbar's number). We could track who helped whom, who cheated, who had what skills.

Modern communities contain thousands to millions of people. Our cooperative instincts are intact but our cooperative perception is overwhelmed. We want to help. We want to reciprocate. We simply cannot see the cooperative opportunities that surround us at scale. The LVN is, fundamentally, a prosthetic for cooperative perception — extending our biological capacity for reciprocity into communities too large for our evolved senses to navigate.

The question isn't "are humans cooperative?" — we demonstrably are. The question is: "what happens when cooperative creatures are embedded in networks too vast for their perceptual systems to navigate?" The answer is what we see around us: willing cooperators, failing to cooperate.

Nowak 2006; Dunbar 1992; Tomasello 2009; Bowles & Gintis 2011

The Core Thesis

The cooperative capacity of humanity vastly exceeds its cooperative performance — and the gap is perceptual, not motivational.

People aren't failing to cooperate because they're selfish, or lazy, or don't care. They're failing to cooperate because the cognitive, social, economic, and institutional systems they inhabit make cooperation invisible. The LVN doesn't try to make people more cooperative. It tries to make the cooperative opportunities that already exist perceptible — using AI as a prosthetic for the social cognition that evolution optimized for 150-person bands, not million-person cities.

The Failure Cascade

Each level compounds the others. You can't fix the gap by addressing any single layer alone.

Cognitive blindness

Experts can't see their own expertise → latent capacity goes unrecognized

Network homophily

We cluster with similar people → complementary matches never form

Market failure

Unarticulated needs can't enter any marketplace → no price signal, no discovery

Institutional gaps

No institution is designed to surface cooperative complementarity → the gap is nobody's job

Scale mismatch

Cooperative instincts evolved for 150 people → modern communities overwhelm perception

The Architecture

Five Layers of Cooperative Infrastructure

The LVN is a protocol, not a platform. It enables AI agents that already understand your situation to discover cooperative matches — without sharing your private data.

Personal Context Engine

Your AI assistant already understands your skills, challenges, and resources through normal conversation. This context stays on your device — sovereign and private.

Latent Capacity Discovery

AI surfaces capabilities you don't recognize in yourself. Resolved three insurance disputes? That's latent negotiation expertise. Explain medical info to your family? That's health communication capacity.

Signal Exchange Protocol

Your agent emits abstract, non-identifying signals — like a lock that only responds to specific key shapes. Complementary agents detect matches without either party exposing private context.

Trust & Reputation Fabric

Distributed trust flows through relational chains — contextual, privacy-preserving, and bootstrappable through small cooperative acts. No centralized trust score.

Governance Layer

Following Ostrom's principles for commons governance. Anti-capture mechanisms built into the protocol. Community-driven evolution with equity monitoring.

The Research

This isn't speculation. It's quantified.

The evidence base spans social network science, labor economics, organizational psychology, and community development research. Across disciplines, the finding is consistent: the cooperative potential that goes unrealized dwarfs what any platform has captured.

83.4%

of jobs found through occasional or rare contacts, not close friends

Granovetter 1973; Rajkumar et al., Science 2022

Cross-class friendships are the single strongest predictor of upward economic mobility

Chetty et al., Nature 2022 (21B friendships)

42%

of organizational skills are known only to the person who holds them

Organizational knowledge loss research

2.5h

per day spent searching for information someone nearby already has

IDC knowledge worker study

76%

of mutual aid requests go unfilled due to matching failures, not lack of capacity

Bender et al., 2024

r=.58

correlation: trust levels predict cooperative behavior across 147 nations

Xue et al. 2025 meta-analysis

Your Path In

We need different minds for different problems

The LVN is deliberately multidisciplinary. Here's where your expertise fits.

For Researchers & Academics

The LVN draws on network science, information economics, mechanism design, and trust research — but several core technical problems remain unsolved. The protocol specification identifies six formal open questions, each of which could be a research contribution. We're particularly looking for people working on privacy-preserving computation, social network analysis, and commons governance.

📄Evidence Base (WD-002)22 citations mapping the gap

🔬Protocol Specification (WD-003)Formal open questions

💬Provocation #2Privacy-preserving semantic matching

For Engineers & Developers

The signal exchange protocol needs implementation. We need people who understand MCP, A2A, distributed systems, and cryptographic primitives. The protocol spec defines data structures, state machines, and integration points — but we haven't written a line of implementation code yet. The architecture is designed to build on existing AI coordination standards.

⚙️Protocol Specification (WD-003)Signal vectors, handshake state machine

🔧GitHub RepositoryIssues, architecture, contribution guide

💬Provocation #8AI assistant interoperability

For Community Practitioners & Organizers

You know what cooperation actually looks like on the ground — the mutual aid networks, the informal help that keeps communities functioning. The LVN needs your reality check. Does this architecture match how cooperation actually works? What are we missing? We're designing pilot programs and need practitioners who can tell us when we're being naive.

📖White Paper (WD-001)The foundational vision

🎯Interactive DemoExperience the matching concept

💬Provocation #10Where should we pilot first?

For Economists & Ethicists

The hardest problems in the LVN aren't technical — they're about incentives, governance, and values. How do you design a value exchange system that accommodates gifts and markets simultaneously? How do you govern a digital commons without recreating the platform capture problem? How do you ensure the system serves the least-connected, not just the most?

📊Evidence Base (WD-002)Quantified gap across 6 domains

💬Provocations #4, #5, #7Equity, pricing, platform capture

🏛️Provocation #6Trust bootstrapping without bias

For Curious Minds

You don't need credentials to contribute. Some of the most important insights come from people who experience the problem firsthand. If you've ever needed help that was nearby but invisible — if you've ever had skills that nobody asked you to use — you understand the Latent Value Gap better than any researcher.

🎯Interactive DemoDiscover your latent value

🤔10 ProvocationsPick one, take a side

💬Provocation #9Is the gap even real?

The Working Documents

Everything is open. Everything is a draft.

These documents define the current state of the initiative. All are working drafts released under CC BY-SA 4.0. Critique is as valuable as contribution.

Foundational White Paper

Defines the 5-layer architecture, 6 design principles, and the core thesis: AI assistants can surface cooperative connections invisible to explicit search.

WD-001

Foundational White Paper

5,000 words defining the vision, architecture, and principles

Complete v0.1

Evidence Base

22 academic citations across 6 evidence streams. Reviews network science, mechanism design, trust formation, and privacy research supporting the approach.

WD-002

Evidence Base

22 academic citations quantifying the Latent Value Gap across 6 domains

Complete v0.1

Signal Exchange Protocol

Layer 3 specification: how context becomes privacy-preserving signals. Bloom filters, differential privacy, progressive disclosure handshake, complementarity scoring.

WD-003

Signal Exchange Protocol

Formal specification for privacy-preserving cooperative matching

Complete v0.1

Explainer Script

2-minute video storyboard with 6 scenes. Takes the concept from abstract to concrete through the story of two neighbors who never knew they could help each other.

WD-004

Explainer Script

2-minute animated storyboard for the concept video

Complete v0.1

Trust & Reputation Fabric

Layer 4 specification: relational, contextual trust. Chain-of-vouching, ZK verification, institutional anchors, Sybil resistance, trust bootstrap mechanics.

WD-005

Trust & Reputation Fabric

Layer 4: distributed contextual trust with ZK verification

Complete v0.1

WD-005

Trust Fabric Spec

Layer 4: distributed contextual trust with ZK verification

Complete v0.1

WD-006

Governance Framework

Layer 5: Ostrom-based commons governance, anti-capture mechanisms

Complete v0.1

The Hard Questions

10 Provocations That Need Your Mind

These are genuine debates where reasonable people disagree. Pick the one that annoys you most — where you think we're wrong — and tell us why.

Should the embedding model be centralized or federated?

Who controls what "complementarity" means controls who gets matched with whom.

Is privacy-preserving semantic matching actually possible?

ZK proofs work for exact attributes. Can they handle fuzzy, semantic complementarity?

Does ambient matching inevitably become surveillance?

A system that understands your needs also understands your vulnerabilities.

How do you prevent the network from only serving the already-connected?

Technology adoption follows power laws. The early adopters aren't the people who need this most.

Should matches include a price?

Does adding value signals make exchanges more equitable — or commodify cooperation?

Read All 10 Provocations →

Common Questions

How is this different from LinkedIn, Nextdoor, or a timebank?

Those platforms require you to explicitly search for what you need. The LVN works in the opposite direction — your AI assistant already understands your context and can discover cooperative opportunities you'd never think to search for. You don't post a listing; connections surface ambientally. Plus, privacy is architectural (your data never leaves your device), not policy-based.

PlatformsYou search for matches.
Your data on their servers.
Profit-driven.

TimebanksYou list services.
Hour-for-hour exchange.
Same community only.

LVNMatches surface to you.
Data stays on your device.
Protocol, not platform.

If my data never leaves my device, how does matching work?

Your AI assistant converts your context into abstract signal vectors — compressed, privacy-protected representations that reveal cooperative potential without exposing private details. These signals go through a community relay that finds complementary matches. If a match is found, you control exactly what to disclose through progressive disclosure: first anonymous categories, then scoped details, then identities — only with mutual consent at every step.

Doesn't this just recreate the "platform capture" problem?

No — and preventing that is a core design constraint. The LVN is a protocol (like email or HTTP), not a platform (like Facebook). Anyone can run a relay. The specifications are CC BY-SA 4.0. Anti-capture mechanisms include term limits on governance roles, forking rights, funding transparency (no single funder >30%), and the Ethics & Equity advisory veto. See WD-006 for the full governance framework.

Why "complementarity" instead of "similarity"?

Similarity matching connects you with people like you. Complementarity matching connects you with people who have what you need and need what you have. This is the fundamental insight: the most valuable cooperative connections cross social boundaries. The retired engineer and the first-generation business owner don't share demographics, interests, or social circles — but they have exactly what the other needs.

What stage is this project at?

Phase 0 (Foundation) is complete — 6 working documents, a cooperative taxonomy, a development plan with 12 work packages, and this website. Phase 1 (Standards & Community) is beginning: we're submitting protocol extensions to MCP and A2A, forming 5 working groups, and identifying pilot communities. Phase 2 (Build) targets June–October 2026. See PLAN.md for details.

How can I contribute?

We need people from many disciplines — not just engineers. Cryptographers, economists, ethicists, community organizers, librarians, social workers, and practitioners all have essential perspectives. Start by trying the demo, reading a provocation that interests you, and then opening an Introduction issue on GitHub or emailing latentvaluenetwork@proton.me. The most valuable contribution right now is telling us what's wrong with our approach.

What's New

Feb 18, 2026

UX Design System + All 20 Audit Items Resolved

Dark mode, animated hero, voting on provocations, match filtering, save/resume demo progress, interactive architecture diagram, social cards

UXDesign System

Feb 18, 2026

Development Plan + Taxonomy + Working Groups

PLAN.md with 12 work packages, cooperative taxonomy v0.1 (170+ entries), 5 working group charters, pilot assessment criteria

SpecsCommunity

Feb 18, 2026

AI Protocol Integration Strategy

Concrete path to standardize cooperative intelligence through MCP SEP, A2A extension, and CIP protocol that composes both

SpecsStandards

Feb 15, 2026

WD-006 Governance Framework

Ostrom's 8 principles applied to protocol governance, anti-capture mechanisms, equity veto, phased evolution pathway

WD-006Governance

Feb 12–14, 2026

Foundation Complete

WD-001 through WD-005, interactive demo, provocations, architecture diagram, GitHub repository with 11 seeded discussions, outreach materials

Core SpecsOutreachDemo

The hardest problems are open by design.

The gap between humanity's cooperative capacity and cooperative performance is the defining infrastructure challenge of our era. We're designing an open protocol to close it — not a company, not a platform. We need cryptographers, economists, ethicists, engineers, organizers, librarians, social workers, and anyone who knows what cooperation looks like on the ground.

The most important thing you can do right now isn't build — it's challenge. Tell us what's wrong with this approach. The protocol gets stronger every time someone finds a flaw.

Join on GitHub Try the Demo Debate the Open Questions View Architecture

Roadmap · Reference Implementation · Research Landscape

Not on GitHub? No problem.

Introduce yourself by email — we want practitioners, organizers, and curious minds, not just developers.

Send an Introduction →

What if the help you need
is already nearby?

The Latent Value Gap

This connection died in silence.

Why can't we see each other?

We don't know what we know — and we can't imagine what others need.

Our networks are silos, and social platforms reinforce them.

Markets are brilliant at pricing — and blind to everything else.

We've confused connection with cooperation.

We're wired for reciprocity — at a scale our biology can't handle.

The cooperative capacity of humanity vastly exceeds its cooperative performance — and the gap is perceptual, not motivational.

Five Layers of Cooperative Infrastructure

Personal Context Engine

Latent Capacity Discovery

Signal Exchange Protocol

Trust & Reputation Fabric

Governance Layer

This isn't speculation. It's quantified.

We need different minds for different problems

For Researchers & Academics

For Engineers & Developers

For Community Practitioners & Organizers

For Economists & Ethicists

For Curious Minds

Everything is open. Everything is a draft.

10 Provocations That Need Your Mind

Common Questions

What's New

The hardest problems are open by design.

What if the help you needis already nearby?

The Latent Value Gap

This connection died in silence.

Why can't we see each other?

We don't know what we know — and we can't imagine what others need.

Our networks are silos, and social platforms reinforce them.

Markets are brilliant at pricing — and blind to everything else.

We've confused connection with cooperation.

We're wired for reciprocity — at a scale our biology can't handle.

The cooperative capacity of humanity vastly exceeds its cooperative performance — and the gap is perceptual, not motivational.

Five Layers of Cooperative Infrastructure

Personal Context Engine

Latent Capacity Discovery

Signal Exchange Protocol

Trust & Reputation Fabric

Governance Layer

This isn't speculation. It's quantified.

We need different minds for different problems

For Researchers & Academics

For Engineers & Developers

For Community Practitioners & Organizers

For Economists & Ethicists

For Curious Minds

Everything is open. Everything is a draft.

10 Provocations That Need Your Mind

Common Questions

What's New

The hardest problems are open by design.

What if the help you need
is already nearby?