Listen Get

Earth Optimization Protocol v1

The current open-source instruction manual for playing the Earth Optimization Game

Abstract

Humans are the contestants in the Earth Optimization Prize. The Earth Optimization Protocol v1 is the current open-source instruction manual we know for helping humanity raise healthy life years and median income within 15 years. It combines eleven mechanisms into a self-reinforcing loop: a 1% military-to-medical treaty provides $27.2B/year in funding; pragmatic clinical trials executed through a decentralized FDA protocol reduce trial costs by 44.1x (95% CI: 39.4x-89.1x) and compress the treatment discovery queue from 443 years (95% CI: 324 years-712 years) to 36 years (95% CI: 11.6 years-77.1 years); Incentive Alignment Bonds fund the lobbying campaign and pay investors 272% annual returns; wishocratic pairwise comparison allocates the 80% research treasury; and an evidence machine (Optimocracy) tells governments which policies actually work. Each component feeds the next. More cures generate popular support, more bondholder income generates political pressure, and more treaty funding generates campaign support for politicians who voted yes. The ratchet only moves in one direction. This paper documents the complete system: what each component does, how they connect, what numbers they claim, and where we think we are wrong. It exists to be forked and improved until humanity has something better to use.

Keywords

war-on-disease, 1-percent-treaty, medical-research, public-health, peace-dividend, decentralized-trials, dfda, dih, victory-bonds, health-economics, cost-benefit-analysis, clinical-trials, drug-development, regulatory-reform, military-spending, peace-economics, decentralized-governance, wishocracy, blockchain-governance, impact-investing

Introduction

Humans are the contestants. This document is a README for one possible way they could win.

Your species has produced thousands of policy proposals, most of which share a structural feature: they describe a pleasant world without describing a mechanism for reaching it. “We should have universal healthcare” is a destination, not a route. “We should end war” is a wish, not an engineering specification. Policy proposals on Earth work like recipes that list ingredients but skip the cooking instructions, then blame the chef when dinner does not materialize.

The Earth Optimization Protocol v1 is different in exactly one way: it is an integrated system where every component creates the conditions for the next component. The treaty funds the trials. The trials generate returns. The returns pay the investors who funded the campaign that passed the treaty. The campaign creates voter demand that makes expanding the treaty easier than repealing it. Remove any one piece and the loop breaks. That is a weakness (harder to build) and a strength (you cannot lobby your way out of a machine whose parts check each other).

We think this plan is terrible. We also think it is better than what your species is currently doing, which is spending 604 (95% CI: 453-894) times more on weapons than on testing which medicines work, and calling this “civilization.”

There is also a clock. Your destructive economy141 (military spending beyond deterrence plus cybercrime) is 11.5% of GDP and growing faster than the productive economy. The Soviet Union collapsed at 15%. At current growth rates, yours hits 25% by 2033 and 50% of GDP by 2040. At that point, half of all economic activity is destructive, so the rational choice for any individual, company, or nation flips from “build things” to “steal things,” because whatever gets created gets looted as fast as it appears. You have a name for places where this already happened. You call them “failed states.” There is no Colombia to flee to when the global economy is the failed state.

The Earth Optimization Prize142 is not this protocol and this protocol is not the contestant. Humans are the contestants. This paper documents the current open-source instruction manual they could use: fork this repository, fix anything idiotic, and improve the manual until it works better on the two terminal metrics. Every component is documented with what it does, what numbers it claims, and what an improved manual still has to solve. Fork it. Improve it. Replace parts that are wrong. We will be genuinely relieved.

The Three-Sentence Version

Sell Incentive Alignment Bonds143 that raise $1B to legally bribe politicians (via Super PACs and lobbying) into signing a treaty144 that redirects 1% of military spending to clinical trials. Of the resulting $27.2B/year, 80% funds trials through your decentralized FDA145,146, 10% pays investors 272% annual returns, and 10% funds a Super PAC that rewards politicians who voted yes. Returns grow in proportion to the treaty percentage, so every billionaire bondholder becomes a permanent lobbyist for expanding it, creating self-sustaining pressure to reallocate from death to life.

How the Pieces Connect

The system has eleven required functions. The first ten form a loop; the eleventh becomes possible once the loop is running:

  1. Incentive Alignment Bonds raise $1B from investors who want 272% returns and billionaires who prefer not dying of horrible diseases.
  2. That money funds the lobbying campaign: Super PACs, scorecards, legislative drafting, and post-office career guarantees. The referendum campaign runs separately on VOTE points (recruit a verified participant, earn a point).
  3. The campaign passes a 1% Treaty147 redirecting $27.2B/year from military budgets to clinical trials. Balance of power unchanged (everyone cuts the same 1%).
  4. The Decentralized Institutes of Health receives the funds and splits them: 80% to medical research, 10% to investor returns, 10% to political incentives.
  5. The research treasury funds your decentralized FDA protocol, running pragmatic trials at $929 (95% CI: $97-$3K)/patient instead of $41K (95% CI: $20K-$120K)/patient.
  6. Wishocracy148 governs how the research treasury is allocated: citizens do ~20 pairwise comparisons in 5 minutes, and the math produces budget weights.
  7. The Evidence Machine (Optimocracy149 + OPG150 + OBG151) tracks which policies actually improve the two terminal metrics across thousands of jurisdictions.
  8. The Political Dysfunction Tax48 and US Efficiency Audit152 identify the waste pools ($101T (95% CI: $83.3T-$191T)/year globally).
  9. The legal architecture (treaty text, Right to Trial153 & FDA Upgrade Act, statutory tools) creates binding force.
  10. This book is the narrative and coalition wrapper that coordinates adoption.
  11. Algorithmic monetary system (Phase 2): once Wishocracy eliminates the politicians who depend on the money printer, replace the Federal Reserve with a shared ledger targeting 0% inflation. The productivity dividend (new money that maintains stable prices as the economy grows) and a transaction tax flow into one pool, allocated entirely by Wishocracy. UBI is one allocation category among many. No Cantillon effect, no wars funded without consent, no IRS, no welfare bureaucracy.

The loop closes: cured diseases generate popular support and economic returns, which fund further treaty expansion, which funds more trials. Each expansion multiplies all three revenue streams (investor, political, research). The system is designed so that the greedier its participants are, the faster diseases get cured.

The Funding Engine

The 1% Treaty

Every signatory nation redirects exactly 1% of its military budget to the 1% Treaty Fund. That is $27.2B/year from $2.72T in global military spending. Balance of power is preserved because every country cuts the same percentage. No country becomes more vulnerable. Your species has done harder things than this: you cut military spending by 87.6% after WW2 and stumbled into the greatest economic boom in history144,147.

The treaty requires only 2 countries to enter into force. Reservations are not permitted (no creative accounting). Article VIII requires unanimous consent plus 10-year notice for withdrawal, because a treaty that can be abandoned next quarter is not a treaty; it is a suggestion with a flag.

The revenue split is 80%/10%/10%: 80% to clinical trials (the point), 10% to investor returns (the bait), 10% to political incentives (the hook). The corruption is capped at 20% and fully transparent. The other 80% goes directly to clinical trials through wishocratic allocation, where nobody with money gets to decide where it goes.

NoteFork Marker

To replace this component, your alternative must redirect at least as much low-value spending to higher-value use, without requiring any participant to become less selfish.

Decentralized Institutes of Health

The DIH is a thin coordination protocol, not an organization. It does three things: receive treaty funds, allocate them via patient subsidies weighted by disease burden (DALYs) and infrastructure decisions via Wishocracy, and verify results. Everything operational is outsourced to competing providers.

The researcher payment formula replaces the entire grant committee apparatus with one equation: subsidy per patient equals DALYs per patient times value per DALY, where value equals total budget divided by global disease burden. Diseases that kill more people get more funding. This is apparently a novel concept on your planet.

No CEO. No board. Every dollar on a public ledger. Anyone can build a competing alternative. The protocol is intentionally minimal to resist capture.

NoteFork Marker

To replace this component, your alternative must allocate research funding at least as efficiently with equal or better capture resistance.

The Medical Engine

Your decentralized FDA

Your species has 9.5 thousand compounds (95% CI: 7 thousand compounds-12 thousand compounds) known safe compounds. 99.7% (95% CI: 99%-100%) of their potential uses have never been tested. At the current discovery rate, clearing the backlog takes ~443 years (95% CI: 324 years-712 years). Your current clinical trial system costs $41K (95% CI: $20K-$120K)/patient, has 1.9 million patients/year (95% CI: 1.5 million patients/year-2.3 million patients/year) slots for 2.4 billion people (95% CI: 2 billion people-2.8 billion people) patients who need them, and excludes 86% of the population.

Your decentralized FDA145,146 is an open protocol (HTTP for clinical trials) that connects treatments to patients, records outcomes, and publishes results. Two stages: observational signal detection at ~$0.1 (95% CI: $0.03-$1)/patient, then pragmatic trial confirmation at ~$929 (95% CI: $97-$3K)/patient. At treaty funding, this expands trial capacity by 12.3x (95% CI: 4.2x-61.4x) and compresses the queue to ~36 years (95% CI: 11.6 years-77.1 years).

This is not theoretical. The Oxford RECOVERY trial tested 6 treatments on 48,000 patients for $500 (95% CI: $400-$2.5K)/patient (versus $41K (95% CI: $20K-$120K) standard), found dexamethasone cut COVID deaths by 30%, and saved 1 million lives (95% CI: 500 thousand lives-2 million lives) in 3 months90,91. Your species does its best medical research when terrified and disorganized.

Products: treatment search and rankings, Outcome Labels (nutrition-facts-style drug effectiveness summaries), real-time safety monitoring. The enabling legislation is the Right to Trial & FDA Upgrade Act, which eliminates the 8.2 years (95% CI: 4.85 years-11.5 years) delay between proving a drug is safe and letting dying humans take it. For every 1 person protected from a dangerous drug, 3.07k (95% CI: 2.88k-3.12k) die waiting for a safe one154.

NoteFork Marker

To replace this component, your alternative must produce faster, cheaper, more reliable treatment discovery, and handle the regulatory-delay problem.

The Political Engine

Incentive Alignment Bonds

The obvious question: who funds a campaign to redirect military spending? Answer: people who want 272% annual returns.

Incentive Alignment Bonds143 fund the lobbying campaign. Buy a bond, your money goes to scorecards, PACs, legislative drafting, and post-office career guarantees for politicians who champion the treaty. If the treaty passes, you receive a pro-rata share of 10% of treaty funding flows: $2.72B/year on $1B total bonds. That is 272% annual returns, paid as a perpetual revenue share for as long as the treaty generates funding.

The secondary market price is a real-time, money-backed estimate of implementation probability. Every bondholder is financially incentivized to recruit voters, lobby politicians, fund implementation work, and share the plan, because all of these raise bond value. Nobody coordinates the lobbying campaign. The bond price coordinates it. And because investing is the dominant strategy regardless of whether a media cascade occurs, the system does not require virality to function.

Three-layer architecture for political adoption: (A) an independent organization publishes Public Good Scores based on politician voting records, (B) PACs and 501(c)(4)s support high-scorers with campaign funding, (C) foundations guarantee post-office careers for champions. No money flows directly to politicians. The same K Street mechanics, pointed at diseases instead of defense contracts. Scoring is based on public voting records; anyone can verify or publish competing scores, making the scoring organization structurally difficult to capture.

The maximum cost of achieving any policy change through legal democratic channels is $1B for the United States, with a benefit-cost ratio exceeding 400,000:1155. “Political impossibility” is a capital allocation problem, not a physics problem.

NoteFork Marker

To replace this component, your alternative must make treaty passage at least as incentive-compatible for selfish actors, with equal or better investor returns.

The Campaign

The campaign has two prongs: a Global Referendum and a lobbying operation.

The referendum proves that the support already exists. Everyone thinks ending war and disease is crazy because everyone else thinks it is crazy. Your economists call this pluralistic ignorance. The referendum is the moment eight billion people discover they nearly all agree. The canonical end-state coordination target in the prize model is 4 billion of people (95% CI: 3.92 billion of people-4.08 billion of people), or 50% of humanity, described in prose as roughly 90% of likely voters globally. More than 10 times as many downloaded TikTok.

The right institutional framing is three lanes, not one giant capital ask:

Lane What it funds Canonical number
Activation underwriting Referral rewards, verification, payment rails, platform ops, launch infrastructure

$30B (95% CI: $15.7B-$46.4B)
PRIZE pool seeding Long-dated success-side upside that makes the scoreboard feel rich and real

$1.8T (95% CI: $1.04T-$3.35T)
Later scale capital The much larger capital migration if the network becomes politically and operationally real $33.7T (95% CI: $1.37T-$286T) (at 1% (95% CI: 0.1%-10%) participation)

The expected first mover is a foundation consortium that can underwrite the stack together while still routing each lane through the committee or vehicle that fits it. One extraordinary institution could still do everything.

$250M is the older narrow-bloc referendum budget for roughly 280 million verified votes. The canonical 50%-of-humanity activation ask is $30B (95% CI: $15.7B-$46.4B). The lobbying budget ($650M) outbids the defense lobby’s annual spend ($127M (95% CI: $100M-$160M)/year) by a factor of 8.

Cash bonuses are the immediate incentive. The longer-dated incentive is VOTE points: verified recruitment credit on the scoreboard and, formally, the success-side claim on the PRIZE pool if the year-15 targets are met. The point is to make coalition-building legible and worth doing, not to create a second speculative asset before the coalition exists.

The Earth Optimization Prize pool is a separate instrument from the bonds. PRIZE depositors place funds into the same pool that compounds until year 15. If global HALE has not reached 79 years (95% CI: 75.7 years-82.3 years) and global median income has not reached $504K (95% CI: $234K-$1.87M) by year 15, depositors divide the realized pool pro rata by deposits. If both targets are met, the same realized pool is distributed to VOTE point-holders proportional to recruitment. A conditional commitment: your money only routes to VOTE point-holders if someone earned it. Philanthropists buy because they want to fund success. Pessimists buy because failure still leaves them with a claim on the pool. Both grow the prize pool. In institutional terms, activation underwriting makes the network real first; PRIZE pool seeding makes the long-dated upside feel rich enough to accelerate it; later scale capital turns the mechanism into a civilization-sized allocator.

NoteFork Marker

To replace this component, your alternative must coordinate adoption at equal or lower cost and failure risk.

Why Every Actor’s Selfish Move Aligns

The system does not require anyone to become a better person. It requires each actor to do the thing they were already going to do, except pointed at diseases:

Actor What they want What the system gives them
Defense contractors Revenue Keep 99% of budget + new health market worth $27.2B/year. The main difference is whether the end user survives.
Pharma Cheaper trials, more revenue Trials become revenue instead of cost; cost drops 44.1x (95% CI: 39.4x-89.1x)
Insurance companies Fewer claims Cured patients stop filing; their lobbyists become your lobbyists
Politicians Reelection + career Super PAC funding + post-office careers proportional to treaty support
Billionaires Returns + not dying 272% returns + their diseases get cured too
Lobbyists Clients + salary Same job, same salary, different client; Wikipedia page improves
Regular humans Health + income Diseases cured + $3.48M (95% CI: $1.05M-$9.82M) lifetime income gain

The Olsonian problem (concentrated interests beat diffuse beneficiaries) is solved by creating concentrated interests on the pro-health side. Bondholders, politicians receiving PAC funding, defense contractors accessing the new health market: all now have per-capita stakes in the treaty’s survival that match or exceed the defense lobby’s per-capita stakes in the status quo.

The Information Engine

The Evidence Machine

Optimocracy149 measures two terminal metrics (median healthy life years and median real after-tax income) across ~10,000 jurisdictions running natural experiments. The Optimal Policy Generator150 turns that data into enact/replace/repeal recommendations. The Optimal Budget Generator151 does the same for spending. Combined: “What works?” answered by data, not argument.

The Political Dysfunction Tax48 is the gap between actual governance output and optimal-governance output: $101T (95% CI: $83.3T-$191T)/year globally. The US Efficiency Audit152 itemizes $4.9T (95% CI: $3.62T-$6.5T)/year across ten categories. These are the waste maps that tell you where the money is.

Why only two terminal metrics? Each additional metric creates a new Goodhart surface. Markets optimize profit, not 200 sub-metrics. Median income is hard to fake at scale. Mortality is binary. Every major policy domain (employment, healthcare, environment, education) flows through income and mortality; what doesn’t affect how long you live or how much you earn probably isn’t worth optimizing. Two numbers. Your species can probably keep track of two numbers.

NoteFork Marker

To replace this component, your alternative must generate better policy and budget recommendations under real-world constraints, and identify low-value spending pools at least as well.

The Governance Engine

Wishocracy

Wishocracy148 asks citizens to drag a slider between two random budget priorities. “How would you split $100 between cancer research and malaria prevention?” About 20 comparisons, 5 minutes. Statistical models (Bradley-Terry/PageRank) aggregate millions of pairwise judgments into budget-preference weights without speeches, lobbying, or horse-trading.

Why pairwise comparisons? Random pairs prevent gaming (no ballot measure to advertise for). Independence is structurally guaranteed (no spoiler candidates, no strategic voting). Outliers cancel out. Your species tried direct democracy four times (Athens, California propositions, Brexit, Swiss referendums) and each failed for a specific reason that pairwise comparison structurally prevents.

In Phase 1, Wishocracy governs how the 80% research treasury is allocated, and infrastructure decisions within the DIH. In Phase 2, it governs the entire public goods budget: the productivity dividend and transaction tax revenue from the algorithmic monetary system flow through Wishocracy, with UBI as one allocation category among many. It is scope-locked to bounded allocation questions. You cannot use it to vote on rights, bodies, or anything unbounded.

NoteFork Marker

To replace this component, your alternative must recover public preferences at equal or lower cognitive cost, with equal or better resistance to capture by organized factions.

The Monetary System (Phase 2)

The central-banks chapter makes the case that the money printer is the root cause: every war that killed more than a million people required one, the hard-money list is empty, and the Cantillon effect has transferred 93% of your purchasing power from workers to whoever stands closest to the printer since 1913. A plan that claims to optimize Earth but ignores the monetary system is a plan that fixes the symptoms and leaves the disease.

V1 does not fix this in Phase 1. The treaty, the bonds, and your decentralized FDA all operate within existing fiat currencies. This is deliberate. Replacing the Federal Reserve requires eliminating the politicians who depend on it, which requires Wishocracy to be running, which requires the treaty to pass, which requires the bonds to sell. You cannot skip to the end.

Phase 2 (post-treaty, post-Wishocracy) replaces the money printer with an algorithmic monetary system described in Wishonia. The core mechanism: a shared ledger targeting 0% inflation. When the economy grows, the algorithm creates exactly enough new currency to keep prices stable. That new money is the productivity dividend, flowing through Wishocracy as the global public goods budget. No congressional budget process. No lobbyist.

A transaction tax (automatic, collected by smart contract) provides the second stream. Both flow into one pool; Wishocracy allocates the entire thing. UBI is one allocation category among many, not a constitutional carve-out. The Optimitron shows the ROI of each option; the aggregate produces a natural equilibrium.

Together, three pieces of software (the monetary algorithm targeting 0% inflation, the transaction tax contract, and Wishocracy) render redundant: the Federal Reserve, the IRS, income tax, the congressional budget process, every welfare program, grant committees, the lobbying industry, and the financial sector.

This is the current manual’s most honest gap. The system that funds wars without democratic consent, steals 2% of purchasing power per year by design, and channels genius toward weapons instead of medicine is not addressed until Phase 2. A revision that solves this earlier, without breaking the Phase 1 sequencing, would be strictly better.

NoteFork Marker

To replace this component, design a transition path from fiat to algorithmic money that does not require the treaty to pass first, or prove that the treaty can pass faster without monetary reform.

The Ratchet

The system is designed to ratchet. Three reinforcing loops ensure the treaty percentage increases but never decreases:

Loop 1: Cures create voters. Every disease cured by treaty-funded trials creates a constituency of formerly dying humans who now vote. Repealing the treaty means telling cured patients “we would like to stop curing diseases now.” No politician survives that sentence.

Loop 2: Returns create lobbyists. Bondholder returns scale with treaty percentage. Every expansion from 1% to 2% doubles investor income. Every bondholder becomes the world’s most motivated lobbyist for expansion. Not because they care about diseases. Because they care about money.

Loop 3: The Super PAC creates career incentives. The 10% political incentive fund rewards politicians proportional to their treaty support. More treaty funding means more PAC money means more politicians competing to be the most pro-treaty candidate.

All three loops compound. The ratchet only turns one way.

The Expansion Math

The treaty money splits 80/10/10: 80% to clinical trials (the point), 10% to investors (the bait), 10% to political incentives (the hook). That last slice grows every time the treaty expands:

Treaty % Political Incentive Pool What Happens
1% $2.72B/year Treaty passes. Humans cautiously try not dying.
2% $5.44B/year Politicians compete to expand (they’ve discovered voters like living)
5% $13.6B/year Health lobby rivals defense lobby. Lobbyists switch suits.
10% $27.2B/year “Funded pragmatic clinical trials” becomes the most rewarded phrase in politics
25% $68B/year Superpowers compete on health outcomes (much less radioactive)
50%+ $136B+/year Historians write confused books about why this took so long

Bond investors get 10% of all treaty funds. At 1%, that is 272% returns. At 2%, it doubles. At 10%, they collect ten times their original payout. Every bondholder calls their senator, not because they care about diseases but because they care about money. You do not have to ask them to do this. They will do it automatically, the way a plant grows toward sunlight except the sunlight is money and the plant is a billionaire. You have built a defense lobby in reverse.

Politicians see their IAB-funded rewards grow with each expansion. By 10%, the political incentive pool ($27.2B/year) exceeds all current global lobbying combined. “I support pragmatic clinical trial funding” becomes the most profitable sentence a politician can say, which means they will say it constantly, which means they will eventually believe it. This is how your species forms convictions.

The Sequencing

You cannot skip steps. All timelines below are order-of-magnitude estimates, not promises.

Phase 1: The Planetary Constitutional Convention156 (weeks to months). Run the constitutional convention. Teams fork the current manual, improve it, and fix the weak parts. Expert panel filters for constitutional compliance. Adversarial review catches what the panel misses. Wishocracy ranks the survivors via pairwise comparison when needed. The ratified revision becomes the current manual. PRIZE deposits can begin during this phase, funding the prize pool regardless of which revision is ratified. This is the cheapest stage and it stress-tests the manual before committing real resources.

Phase 2: The referendum (weeks to months). Launch PRIZE deposits and the referendum platform simultaneously. Each participant votes on the 1% Treaty (the headcount that creates political pressure) and sets a wishocratic slider for their preferred military-to-health allocation (the data that tells you what people actually want). Every recruiter who brings a verified participant earns a VOTE point. The institutional financing pitch is three-lane: underwrite activation first, seed the PRIZE pool second, and invite much larger capital once the network is visibly real. The operational question is not whether the network can reach some arbitrary macro benchmark. It is whether activation capital can make participation self-propagating before the later pool and capital-allocation thesis take over.

Phase 3: Pass the treaty (months to years). Scale the referendum toward 4 billion (95% CI: 3.92 billion-4.08 billion) verified supporters. Launch the full VICTORY Incentive Alignment Bond sale ($1B target). Overwhelm the defense lobby ($127M (95% CI: $100M-$160M)/year) with 8x their spend. Pass the treaty. $27.2B/year begins flowing.

Phase 4: Full optimization (years to decades). Treaty expands from 1% to 2% to 5% to 50%+. Each expansion multiplies all three revenue streams. Wishocracy governs how the research treasury is allocated. Once Wishocracy is running at scale, replace the Federal Reserve with the algorithmic monetary system. A transaction tax replaces the IRS. UBI becomes one wishocratic allocation category among many. Three pieces of software replace the entire fiscal and monetary apparatus.

The Numbers

The ROI is astronomical because three compounding engines run simultaneously. Engine 1: R&D spillover. Redirect engineers from missiles to medicine. Every dollar of R&D produces 2x (95% CI: 1.5x-2.5x) in economy-wide productivity gains; military R&D produces approximately none (classified, non-transferable, designed to destroy). Engine 2: Disease burden reduction. Disease currently costs 13% of GDP in sick workers, medical bills, and lost productivity. Cure the diseases, recover the GDP. Engine 3: Governance improvement. The Optimitron identifies $101T (95% CI: $83.3T-$191T)/year in political dysfunction. Fix even a fraction and the gains compound. All three run for 20 years. Cut every assumption in half, then in quarters. The model still produces a civilization instead of a collapse.

The economic case, compressed:

Metric Value
Treaty funding $27.2B/year
Trial cost reduction 44.1x (95% CI: 39.4x-89.1x) (from $41K (95% CI: $20K-$120K) to $929 (95% CI: $97-$3K)/patient)
Trial capacity increase

12.3x (95% CI: 4.2x-61.4x)
Disease queue 443 years (95% CI: 324 years-712 years) compressed to 36 years (95% CI: 11.6 years-77.1 years)
Lives saved

10.7 billion deaths (95% CI: 7.4 billion deaths-16.2 billion deaths)
Economic value

$84.8 quadrillion (95% CI: $62.4 quadrillion-$97.3 quadrillion)
Campaign cost

$1B
ROI 84.8M (95% CI: 46.6M-144M) to 1
Cost per DALY $0.00177 (95% CI: $0.000715-$0.00412) (bed nets: $89 (95% CI: $78-$100))
Investor returns 272% annually (perpetual revenue share)
PRIZE token failure branch Depositors divide the realized pool pro rata in 2040 (PRIZE stays in the pool; IABs are spent on the lobbying campaign)
Lifetime income gain $3.48M (95% CI: $1.05M-$9.82M) per person
GDP multiplier (year 20) 4.88x (95% CI: 2.3x-10.8x) (treaty-only) to 56.7x (95% CI: 19.3x-304x) (full optimization)

Full economic modeling: GDP Trajectories. Full parameter documentation: Parameters & Calculations.

What We Think Is Wrong

Your policy papers never have a section called “Where We Think We Are Wrong.” This one does.

These are the failure modes we consider most likely, ordered by severity. If you are improving the manual, start here.

1. The First Billion

Everything depends on raising $1B in initial IAB capital. If no billionaire buys in, the cascade never starts. The chain reaction model says that even with 90% (95% CI: 80%-97%) dismissal, approximately 3.48 (95% CI: 0.0736-53.7) of 2.98 thousand billionaires and heads of state engage within 3 years. The math is reassuring. The math has not met your billionaires. This is the single highest-risk node in the system.

How to improve this part: PRIZE tokens keep philanthropists and pessimists in the same pool: failure leaves depositors with a pro-rata claim on the realized pool, while success routes the same pool to VOTE point-holders. IAB investors take real risk (money funds the lobbying campaign), but the revenue share on success (272% annually) makes the expected value positive even at low success probabilities. The decision-theoretic case shows investing is the dominant strategy even without a cascade. What neither proves is that real billionaires evaluate payoff matrices. A pilot at $10M that demonstrates the loop before asking for $1B would reduce this from a modeling question to an empirical one.

2. The 20% Overhead Question

The 80%/10%/10% split is a design choice, not a theorem. The combined overhead (10% investor returns + 10% political incentives) looks terrible to philanthropists who expect 95% program ratios. It looks excellent to investors who expect 10% returns. Whether the investor-attraction benefit exceeds the philanthropist-repulsion cost is an empirical question nobody has answered. The split may be wrong.

How to improve this part: Prove that a different split raises more total capital or produces better outcomes per dollar. Or prove that the 20% is unnecessary because a cheaper political mechanism exists.

3. Wishocracy Participation

If participation in wishocratic allocation is low, the allocations are noisy and manipulable by organized factions. The system requires thresholds, weighting, or fallback modes for low participation. We have specified these. We have not tested them.

How to improve this part: Prove that a different allocation mechanism produces better outcomes with lower participation requirements, or design stronger participation incentives.

4. No Pilot

We have not tested any of this as an integrated system. The RECOVERY trial proves pragmatic trials work. The Montreal Protocol proves coordinated reduction works. The NRA proves political scoring works. But the integrated loop has never run. Individually validated components can fail when combined, the same way individually safe chemicals can explode when mixed.

How to improve this part: Run a pilot. A single country, a single disease category, a scaled-down version of the full loop. Prove the integration works before asking for planetary adoption.

What Already Exists

This is not a whitepaper. The core infrastructure is built and tested:

Package What it does Status
Optimitron core Domain-agnostic causal inference engine 176 tests
Wishocracy RAPPA pairwise preference aggregation, citizen alignment scores 162 tests
Optimal Policy Generator Enact/replace/repeal recommendations from cross-jurisdiction data 213 tests
Optimal Budget Generator Diminishing-returns budget optimization 146 tests
Data layer OECD, World Bank, FRED, WHO, Congress.gov fetchers + 9 health importers 296 tests
Treasury contracts WishToken ($WISH), VOTE tokens, PRIZE tokens, IAB vault, prize pool, alignment treasury (Solidity) Hardhat tests
Web app Live RAPPA surveys, politician scorecards, bill voting, alignment reports Next.js, deployed

Total: 2,600+ tests across 13 packages. TypeScript strict mode. Zod validation throughout. Every algorithm backed by a published paper with exact formulas and worked examples.

When you fork the current manual, you are forking working code, not a PDF. The repository contains every calculation, every parameter, and every assumption. The Optomitron repository contains every algorithm.

How To Improve This Manual

The Earth Optimization Protocol v1 is the current instruction manual for the Earth Optimization Prize. If you think it is wrong, fork it and improve the manual:

  1. Fork. Clone the repository. Every component, every calculation, every assumption is public.
  2. Improve any row in the Minimum Acceptable Governance table (see the prize paper) with something strictly better on the two terminal metrics (median healthy life years and median real after-tax income).
  3. Or replace the wrong parts with an integrated alternative that improves the current manual function by function. You cannot skip inconvenient pieces; every required function must still be covered or replaced.
  4. Prove it. The credit mechanism is VOTE points: verified recruitment credit on the scoreboard and, formally, success-side claims on the PRIZE pool. You need numbers, not arguments.
  5. Use the better manual. Bondholders receive their revenue share for as long as the treaty generates funding. VOTE point-holders receive their share of the PRIZE pool when both global targets are met at year 15 (HALE 79 years (95% CI: 75.7 years-82.3 years), median income $504K (95% CI: $234K-$1.87M)).

The disqualification rule: if your proposal requires any participant to become less selfish, you have described a pleasant world rather than a mechanism for reaching it. The math must work when humans are greedy, impatient, and exactly as self-interested as they are right now.

We call this revision v1 because repositories need version tags. The real goal is to keep improving the current manual until humanity has something better to use.

We expect parts of this manual to be improved or replaced. Good.

Who Does Each Job

The system has eleven functions but only five organizations that someone has to actually be. The rest is software, protocol rules, and compound interest. Here are the five, why each exists, and why none of them can be the same entity as any other.

The Five Roles

1. The Starter (Launch Host). Makes the first public version of the Earth Optimization Prize real. Publishes the challenge rules, convenes the first expert review panel, funds adversarial-review bounties, and launches a live public scoreboard. This is the role that turns a stack of papers into inspectable public infrastructure. It is time-bounded by design: the Starter hosts the first 90-day cycle and then hands off or continues, but at no point becomes the permanent root of trust. One institution may start the network. No institution should be able to stop it. Redundancy model: sequential, not parallel. During bootstrap, one institution must go first or nobody goes at all (this is the current state; diffused ownership is why the prize is still a PDF). After the first cycle, co-hosts and successor hosts can join. Clear ownership during each bounded period, but replaceable between periods. (Full specification: The Initiating Node Memo and the 90-Day Pilot.)

2. The Vault (Pool Custodian). Holds PRIZE deposits, executes the published investment policy, and disburses funds when the oracle triggers release. This is custody, fiduciary duty, and independent audits. The Vault does not decide whether the targets were met. It does not lobby for the treaty. It does not know who the voters are. It holds money, invests it according to published rules, and moves it when told to by the Referee. That is all. Redundancy model: many parallel pools. The protocol treats multiple pools as complements, not competitors. Japan launches a pool, the EU launches a pool, the Gates Foundation launches a pool. Each pool has ONE custodian (clear ownership, someone to audit, someone to sue), but the system has MANY pools. A voter claims proportionally from every participating pool. Launching a “competing” pool is indistinguishable from donating to the network. A bad custodian loses depositors to a better pool. Markets handle discipline. (The custodian role is described in Decentralized Institutes of Health and the Legal Architecture.)

3. The Referee (Measurement Authority). Locks the baseline measurement methodology before the game begins, reads WHO and World Bank data at specified intervals, publishes cryptographically signed determinations of whether the graduated release thresholds have been met, and runs the dispute resolution process when someone challenges the data. The Referee decides whether the targets were hit. The Vault holds the money. These two jobs exist in separate organizations for the same reason you do not let the scorekeeper also be the team owner. Redundancy model: one accountable consortium, verified by many. The inputs are already multi-node (five independent data sources that produce the same answer if nobody is lying). The binding determination has to converge on one signed statement, but the computation is deterministic from public data, so anyone can verify it independently. If the Referee’s determination disagrees with what the raw data says, the bonded challenge mechanism lets anyone post evidence and trigger independent review. The Referee is a single accountable node that is continuously auditable by everyone. (The measurement system is described in Optimocracy.)

4. The Bouncer (Identity Provider). Verifies that each referendum participant is one real human, not a bot, deepfake, or the same person registering twice under a different hat. Maintains or contributes to the global nullifier set (the shared log that prevents double-registration across pools). Without one-person-one-vote enforcement, VOTE points are meaningless and Sybil attacks transfer wealth from legitimate voters to attackers. The Bouncer cannot be the same organization as the Vault, because the entity that knows who people are should not be the entity that holds their money. Redundancy model: start with one, add more with bridging. The protocol defines a provider-agnostic interface so additional providers can be added without re-architecting. Multiple Bouncers add geographic resilience (if World ID gets banned in a jurisdiction, BrightID or government e-ID picks up). The shared state (global nullifier set) is append-only and public, so no single provider’s disappearance destroys the voter registry. Each provider owns its verification process clearly; the protocol handles cross-provider deduplication.

5. The Sword (Campaign Executor). Converts scoreboard support into political action. Raises campaign capital via Incentive Alignment Bonds, operates the SuperPAC that funds treaty-aligned politicians, and coordinates the legal path from “public support exists” to “treaty is signed.” The Sword cannot be the same organization as the Referee, because the entity lobbying for the outcome cannot be the entity measuring whether the outcome occurred. It cannot be the same organization as the Vault, because securities law, election law, and tax law all require political funds and investment funds to stay separated. Redundancy model: many parallel Swords, one per jurisdiction. You need a different legal entity in every country anyway (US 501(c)(4), EU equivalent, etc.). Multiple Swords compete for bondholder capital to run the most effective campaign in their territory. Each Sword owns its jurisdiction clearly; its revenue depends on treaty passage in that territory, so the incentive to perform is direct. A bad Sword loses bondholder capital to a more effective one in the same jurisdiction. No tragedy of the commons because ownership is geographic. (The Legal Architecture describes the four-entity structure; the campaign mechanics are in Incentive Alignment Bonds.)

Why Five and Not Fewer

Four firewalls force the minimum to five:

Firewall Why it exists
Vault/Referee The entity holding the money cannot decide whether to release it. Otherwise it releases the money to itself.
Bouncer/Vault The entity that knows who people are cannot hold their money. Otherwise it coerces voters or sells identity data.
Sword/Referee The entity lobbying for the outcome cannot measure whether it occurred. Otherwise it declares victory whenever convenient.
Sword/Vault Political funds and investment funds stay separated. The IRS, SEC, and FEC all require this. Your species did not invent these firewalls because it enjoys paperwork. It invented them because the alternative was tested and the alternative was stealing.

Merge any two roles that share a firewall and you create the structural vulnerability that a skeptical institutional decision-maker will immediately identify and reject. The five roles are not a wish list. They are the minimum set of independent actors that prevents the system from robbing itself.

Why Not Just Decentralize Everything

Your species has two failure modes for coordination. In the first, one organization does everything, gets captured, and the system robs itself. In the second, nobody is responsible for anything, everyone assumes someone else will do it, and nothing happens. Your economists call the second one the tragedy of the commons. On Wishonia, we call it “a meeting.”

The system avoids both by matching each role to the redundancy model that fits its function:

Role Redundancy model Why
Starter Sequential (one at a time, replaceable between cycles) Someone must go first. Diffused ownership during bootstrap is why the prize is still a PDF.
Vault Many parallel pools (competitive) Each pool has one clear custodian. Markets discipline bad ones. Launching a rival pool helps the network.
Referee One accountable consortium, verified by everyone The determination must converge, but the data is public and the computation is deterministic. Anyone can check the Referee’s work.
Bouncer Start with one, add more with bridging The interface is provider-agnostic. Geographic redundancy adds resilience. The shared nullifier set survives any single provider.
Sword Many parallel Swords (one per jurisdiction) Legal entities are jurisdiction-specific anyway. Each Sword’s revenue depends on its own territory.

The principle: every role has ONE accountable node at any given scope (one custodian per pool, one Sword per jurisdiction, one Referee per determination), but the system has MANY such nodes, so no single failure kills the network. Clear ownership prevents the tragedy of the commons. Replaceability and competition prevent capture. Your species already does this with banks (many banks, each regulated, each replaceable, each clearly responsible for its own customers). It just hasn’t applied the pattern to ending war and disease.

What Is Not a Role

The Scoreboard is open-source software, not an organization. Since the computation is deterministic (aggregate pool balances and voter counts from published endpoints), any two honest aggregators produce identical results. Anyone can run one. The Starter launches the first one; after that, it replicates.

VOTE and PRIZE token mechanics are protocol rules, implementable as smart contracts or as traditional custodial accounting (the protocol specification is deliberately implementation-agnostic). The Vault operates them as part of custody. Not a separate organization.

The adversarial reviewers and red-team critics are participants hired by the bounty pool, not a standing institution.

Building the referendum app, the Optimitron, and the Wishocracy interface are development projects. Much of it already exists. Open source. Not an ongoing organizational responsibility.

Candidates and Self-Nomination

If your organization fits one of these profiles, the next step is a conversation: [email protected].

Role Org type Examples
The Starter A foundation or think tank with existing prize-program infrastructure and credibility in long-term civilizational risk Foresight Institute, X Prize Foundation, Santa Fe Institute
The Vault A regulated financial custodian or purpose-built Swiss foundation with independent auditor Qualified custodians, Swiss foundations with locked charters
The Referee An academic consortium or statistical panel with expertise in health economics and independence from political pressure University health economics departments, WHO-affiliated research groups
The Bouncer A proof-of-personhood service with global reach and privacy-preserving verification World ID, BrightID, Proof of Humanity
The Sword A political advocacy organization with experience in legislative campaigns and compliance infrastructure Purpose-built 501(c)(4), modeled on existing effective advocacy structures

These are illustrative, not exclusive. If your organization can do one of these jobs and wants to, add yourself to the candidate list or email [email protected].