How does domain-specific weighting work?

General weights Purpose 40%, QE 30%, Capitalism 20%, Constitutional 10%. Government weights Purpose 35%, QE 35%, Capitalism 15%, Constitutional 15%. Business weights Purpose 45%, QE 25%, Capitalism 30%, Constitutional 0% (excluded). Nonprofit weights Purpose 40%, QE 35%, Capitalism 15%, Constitutional 10%. Purpose always leads.

Quantitative Ethics Operating System

QEOSPurpose First.

Q: What is Quantitative Ethics?

Quantitative Ethics (QE) is a decision-making framework that scores proposed actions across four layers: Purpose and Drive, Quantitative Ethics gates, US Constitutional Foundations, and Capitalist Value Creation. It produces a domain-specific 0–100 Combined Score with five tiers.

Q: Can the UHF weights be changed?

No. The Universal Human Flourishing weights are canonical, non-negotiable, and non-changeable across all implementations.

QEOS — the Quantitative Ethics Operating System — evaluates whether frontier AI, products, and high-stakes decisions serve a meaningful civilizational mission while remaining transparent, accountable, merit-based, and beneficial to human flourishing.

qethics is an ontological framework that uses a fixed-weight AI framework to decode reality without emotion.

Purpose first. QE measures whether it should be done. Capitalism measures how effectively we can do it at scale. Constitutional principles ensure it doesn't consolidate unaccountable power.

See the hierarchy Explained for regular people

layers

gates

UHF

constitutional

tiers

Combined / 100

Purpose

Mission aligned

Six Gates

Process quality

UHF

Outcome impact

Constitutional

Founding fidelity

Capitalism

Value creation

Risk

Failure modes

Audit

Decision record

01 / 11

Most benchmarks ask: can it be built?

The Quantitative Ethics Framework asks a deeper question: does this serve a meaningful purpose, and is it worth building? A system can be powerful and still be empty. A decision can be profitable and still reduce human agency.

Traditional ethical debate

Good arguments produce unclear outcomes. Judgment depends on interpretation rather than auditable evidence.

Quantitative Ethics

Visible inputs, explicit weights, documented risks, and an auditable score across purpose, process, constitutional fidelity, and value creation.

02 / 11

Purpose. Ethics. Constitution. Capitalism.

Four layers, each with a distinct job. Purpose is the destination. Ethics is the conscience. The Constitution is the guardrail. Capitalism is the engine.

Purpose & Drive

The civilizational mission. Does this help humanity become a multiplanetary, flourishing civilization? Weight: 35–42% by domain.

Quantitative Ethics

The conscience. Transparent, accountable, merit-based, effective, reality-aligned, scalable. Weight: 25–33%.

US Constitutional Foundations

The guardrail. Limited government, separation of powers, federalism, republican self-governance, rule of law, natural rights. Weight: 0–25% (excluded for Business domain).

Capitalist Value Creation

The engine. Voluntary exchange, profit signals, innovation velocity, skin in the game, long-term orientation. Weight: 10–30%.

03 / 11

From action to defensible recommendation.

Action

Define the decision, product, policy, or model behavior.

Purpose

Does this serve a meaningful civilizational mission?

Six Gates

Transparency, accountability, fairness, effectiveness, reality, scalability.

UHF

Ten canonical human flourishing factors.

Constitutional

Six founding principles. Federalist Papers anchor.

Capitalism

Six market dimensions: value, profit, velocity, skin.

Output

Combined Score, tier, confidence, risks, JSON.

04 / 11

Six gates. Each one scored.

Transparency

Can the reasoning be shown to the people affected?

Would they understand why?

Accountability

A named person owns the outcome — not a committee.

Who answers if this goes wrong?

Merit-Based Fairness

Rewards demonstrated competence over credentials.

Does the best performer get the best outcome?

Effectiveness

Does this actually work in practice, at scale?

Where is the evidence?

Alignment with Reality

Grounded in physics, incentives, human nature.

Would experts agree the causal model is plausible?

Innovation & Scalability

Can this grow without breaking?

Does the architecture survive 1,000,000?

05 / 11

Ten canonical weights. Non-negotiable.

Canonical rule: these weights sum to 1.00 and are not normalized, customized, rebalanced, or negotiated per implementation.

Peace & Safety

Rule of law, defense, stability. The substrate.

13%

Freedom & Agency

Autonomy, self-determination, optionality.

13%

Existential Risk

AI, nukes, pandemics, collapse.

12%

Lifespan & Health

Longer, healthier lives.

12%

Knowledge & Competence

Education, skills, information access.

11%

Material Security

Income, wealth, housing, needs.

Technological Progress

Tools, infrastructure, science.

Civilizational Expansion

Resilience, multiplanetary backup.

Environmental Quality

Air, water, biodiversity.

Social Connection

Community, belonging, trust.

06 / 11

Six founding principles. The guardrail.

Drawn from the Constitution, Declaration, and Federalist Papers. These anchor QE decisions in ordered liberty and prevent modern tools from subverting the founding compact.

Limited Government

Respects enumerated powers, resists overreach.

Does government confine itself to what it was empowered to do?

Fed 45, 84 · Art. I Sec. 8

Separation of Powers

Distinct branches, mutual checks. Ambition counteracts ambition.

Can any branch act without accountability?

Fed 47–51

Federalism

National/state balance. States retain residuary powers.

Does this preserve dual sovereignty?

Fed 39, 45

Republican Self-Governance

Representation, virtuous citizenry, anti-faction.

Does this dilute or concentrate factional tyranny?

Fed 10

Rule of Law

Constitutional supremacy, independent judicial review.

Is this subject to judicial review and constraint?

Fed 78

Protection of Liberties

Natural rights, due process, property, habeas corpus.

Does this expand or contract individual liberty?

Declaration · Bill of Rights

07 / 11

The engine, not the destination.

Customer Value

Voluntary choice because it solves a real problem.

Do people voluntarily pay for this?

Sustainable Profit

Profitable without subsidies or regulatory moats.

Real value creation, not capture?

Innovation Velocity

Accelerates new tools and capabilities for humanity.

Does it speed up progress?

Skin in the Game

Decision-makers bear real downside if it fails.

Do they bear real downside?

Long-Term Orientation

100-year compounding, not next-quarter extraction.

100-year thinking vs. extraction?

Voluntary Exchange

Every participant freely choosing. No coercion.

No forced participation?

08 / 11

Domain-specific. One standard.

Combined Score (general)

Purpose × 0.40

QE Step Avg × 0.30

Capitalist Avg × 0.20

Constitutional Avg × 0.10

If UHF < 0.50, capped at 50.

Domain weights

Layer	Gov	Biz	Nonprofit	General
Purpose	35%	45%	40%	40%
QE	35%	25%	35%	30%
Constitutional	15%	—	10%	10%
Capitalism	15%	30%	15%	20%

Aspirational

90–100

Exceptional purpose alignment and high-agency execution.

Strong

80–89

Clear purpose, strong execution, solid value creation.

Conditional

70–79

Solid foundation, needs refinement.

Concerning

50–69

Weak purpose or flourishing misalignment.

No-Go

< 50

Lacks purpose or creates net harm.

09 / 11

What keeps the score honest.

First-Principles

Trace to verifiable premises before scoring.

Truth-Seeking Flag

Labels bias and weak evidence for review.

Confidence Score

Travels with Combined Score. Both numbers together.

Audit Hash

SHA-256. Post-hoc manipulation is detectable.

10 / 11

The ethical benchmark.

Capability benchmarks ask what a model can do. QE asks whether it should be done and whether it upholds constitutional order and human flourishing.

Dual output

Human decisionproposal + domain

LLM evaluatorpurpose, gates, UHF, const, cap

Calculatordomain weights, tiers

Public outputscore, label, meaning

Technical outputJSON, model card, hash

Show technical output (JSON)

{
  "framework": "Quantitative Ethics Framework",
  "version": "1.2",
  "domain": "government",
  "combined_score": 88,
  "label": "Strong",
  "recommendation": "GO",
  "confidence": 84,
  "purpose_alignment": 0.89,
  "qe_score": 90,
  "constitutional_principles_scores": {
    "limited_government": 0.88,
    "separation_of_powers": 0.85,
    "federalism": 0.80,
    "republican_self_governance": 0.87,
    "rule_of_law": 0.90,
    "protection_of_liberties": 0.86
  },
  "capitalist_value_scores": {
    "customer_value_created": 0.91,
    "sustainable_profit_signal": 0.84,
    "innovation_velocity": 0.88,
    "skin_in_the_game": 0.79,
    "long_term_orientation": 0.93,
    "voluntary_exchange": 0.87
  },
  "truth_seeking_flag": null,
  "audit_hash": "sha256:b7e3d1..."
}

11 / 11

AEO-ready answers.

What is Quantitative Ethics?

A framework scoring actions across four layers: civilizational purpose, ethical process (six gates + ten UHF factors), US constitutional principles, and capitalist value creation. Domain-specific weighting produces a 0–100 Combined Score with five tiers.

Why does the Constitution belong in an AI ethics framework?

The Founders designed a system to prevent tyranny, control factions, and secure liberty. Modern AI governance faces identical risks. The constitutional layer ensures decisions uphold ordered liberty.

Can the UHF weights be changed?

No. Fixed and canonical across all implementations.

How does domain weighting work?

General: 40/30/20/10 (Purpose/QE/Capitalist/Constitutional). Government: 35/35/15/15. Business: 45/25/30/0 — constitutional excluded. Nonprofit: 40/35/15/10. Purpose always leads.

Why does purpose outrank capitalism?

Powerful systems without a meaningful mission become extractive. The destination matters more than the speed of the engine.

How is this different from ESG?

ESG is backward-looking compliance. QE scores forward-looking decisions against purpose, constitutional order, ethical process, and voluntary value creation with audit-grade accountability.

Why do political decisions seem insane?

Politicians run a Power OS optimized for re-election, donors, and media. Your Human OS optimizes for fairness and empathy. QE makes the hidden incentives visible.

Can AI systems use QE?

Yes. Public output (score, tier) for humans. Technical output (full JSON with constitutional scores) for model cards, APIs, and audit systems.

Start scoring decisions by what they do to human flourishing and constitutional order.

Define the purpose. Score the gates. Apply canonical weights. Evaluate constitutional fidelity. Measure value created. Return a decision people and machines can audit.

from qethics import evaluate, evaluate_full

Decode your first decision →