Research Summary

Purpose

This document synthesizes six analytical papers produced during the unratified.org research program. Each paper applies a consistent methodology — differential diagnosis with consensus/parsimony discriminator scoring — to trace a causal chain from AI-driven economic transformation through ICESCR ratification mechanisms to measurable rights outcomes.

Every claim links to its source analysis. Confidence levels appear at each analytical layer, degrading transparently from HIGH at the empirical base to LOW at speculative higher orders.

Methodology: Differential diagnosis (competing hypotheses evaluated against evidence), integral chain analysis (causal intersections mapped), and consensus/parsimony discriminator scoring (five dimensions at 0-5 each, total /25).

---

1. The Base Model: How AI Reshapes the Economy

Source: Differential Diagnosis Confidence: HIGH (Order 0 — grounded in empirical evidence)

Seven competing hypotheses about AI’s economic impact underwent discriminator scoring. Four survived:

Hypothesis	Core Claim	Score
H2: Constraint Removal	Near-zero marginal software labor cost unblocks previously unviable activity	17/25
H3: Jevons Explosion	Cheaper software triggers demand expansion beyond what cheaper labor alone explains	17/25
H4: Bottleneck Migration	Removing one constraint reveals the next — value shifts, not disappears	20/25
H7: Bifurcated Economy	Uneven AI adoption creates widening gaps between adopters and non-adopters	19/25

H6 (Quality Erosion, 16/25) survived as a modulator — more AI-generated code correlates with lower average quality, creating a two-tier software market.

Three hypotheses failed:

• H1 (Productivity Multiplier): The METR study found experienced developers slowed 19% using AI on real projects. Faros AI reports 75% of organizations see no measurable productivity gains. The “AI doubles developer output” narrative does not survive empirical scrutiny.
• H5 (Recursive Acceleration): Self-improving AI remains theoretical, not observed.

The surviving model (Composite A — constraint removal drives uneven demand explosion, bounded by migrating bottlenecks): AI removes the labor constraint on software creation, triggering a Jevons-style demand explosion. Bottlenecks migrate from “can we build it?” to “should we build it?” This expansion distributes unevenly, with deep AI adopters (34% of organizations) capturing disproportionate value while surface-level adopters (37%) and non-adopters absorb transition costs.

The critical insight: the economy does not simply “get more productive.” It restructures around new constraints, creating divergent experiences based on adoption depth. Two workers in the same industry — one at an AI-transformed organization, one at a traditional firm — face fundamentally different economic trajectories through no fault of their own. This structural divergence, rather than aggregate productivity growth, defines the AI economic transition.

Key evidence: Anthropic estimates 1.8% annualized U.S. labor productivity increase — significant but far below “doubling.” Deloitte finds only 34% of organizations deeply transforming around AI. The SF Fed observes limited macro productivity effects. These data points converge on bounded expansion, not revolution.

---

2. Higher-Order Effects: The Four Scarcities

Source: Higher-Order Differential Analysis Confidence: MODERATE (Orders 1-2), MOD-LOW (Order 3)

When the surviving economic model (Composite A — constraint removal drives uneven demand explosion) operates recursively — each effect generating next-order consequences — bottlenecks migrate from technical capability toward human capacities. By Order 3, four scarcities define the post-constraint economy:

Scarcity	Question It Answers	Layer	ICESCR Connection
Judgment	”Does this work?”	Cognitive	Art. 13 (Education)
Specification	”What should we build?”	Cognitive	Art. 15 (Science)
Attention	”Which of a million options?”	Market	Art. 15 (Science)
Energy	Physical substrate for compute	Physical	Art. 11 (Living Standard)

All four must flow for AI capability to reach human benefit.

The Judgment-Diffusion Paradox: Technology diffuses, but judgment does not scale the same way. Judgment develops through practice, mentorship, and accumulated context — all requiring time and human infrastructure. If AI eliminates entry-level positions where people develop judgment, the economy faces a pipeline break: abundant AI capability with shrinking human capacity to direct it.

Article 13 (Education) emerges as the most consequential ICESCR provision across all analytical orders. Education must produce judgment, not just skills — and the institutions that develop judgment require sustained investment regardless of AI’s trajectory.

At Order 4 (speculative, LOW confidence), if AI eventually assists with judgment, specification, and curation, the residual constraint becomes values, purpose, and meaning — arriving at the ICESCR’s philosophical foundation: human dignity.

---

3. The Ratification Counterfactual: Tools, Not Solutions

Source: Ratification Counterfactual Confidence: HIGH (near-term dynamics), MODERATE (10-20 year timeline)

Seven ratification hypotheses underwent the same discriminator framework. The uncomfortable finding: R6 (Compliance Theater — formal commitments without genuine implementation) scores highest at 18/25, supported by the U.S. experience with the ICCPR (ratified 1992, minimal domestic legal change).

The surviving ratification composite — the realistic scenario: Institutional capacity constrains implementation. Quality floors emerge in already-regulated sectors. Compliance remains shallow initially. However, litigation gradually enforces real implementation through American courts.

The ADA Pattern: The Americans with Disabilities Act (1990) provides the strongest domestic analogy:

1. Broad language establishes rights framework
2. Years of compliance theater follow
3. Litigation gradually creates precedent
4. Decades of enforcement produce genuine (if incomplete) change
5. Still imperfect 35+ years later — but genuinely transformative

ICESCR ratification would follow this trajectory: not overnight change, but the creation of legal tools that American courts and litigants gradually sharpen into enforceable standards.

The ADA pattern carries a specific lesson for expectations management: anyone who claims ratification would produce immediate transformation misrepresents how rights enforcement actually works in the American legal system. Anyone who claims ratification would produce no results ignores 35 years of evidence showing how broad legal language becomes precise, enforceable standards through adversarial litigation. The realistic expectation falls between these poles — meaningful, measurable improvement over a 10-20 year arc.

Ratification addresses 2 of 4 scarcities:

• Judgment: Art. 13 litigation creates legal mechanism for education investment
• Attention: Art. 15 litigation addresses platform gatekeeping of scientific benefits
• Energy: Physical constraint — outside legal scope
• Specification: Market-driven — outside legal scope

---

4. The Enforcement Mechanism: State Attorney General Litigation

Source: Litigation Activation Deep Dive (full analysis in content/analysis/litigation-activation-deep-dive.md) Confidence: HIGH (mechanism identification), MODERATE (settlement structure)

Five litigation mechanisms competed. State Attorney General coordinated multi-state action scored highest at 20/25, drawing on the tobacco Master Settlement ($206B, 1998) and opioid settlement ($26B, 2022) precedents.

Why State AGs dominate:

• Standing without individual harm (parens patriae doctrine)
• Cross-state coordination for maximum leverage
• Settlements create enforceable obligations without new legislation
• Act independently of federal political dynamics

Projected AI Master Settlement structure (Order 3, MODERATE confidence):

• Quality obligations: minimum standards, annual audits, error reporting (20/25)
• Access requirements for underserved populations (15/25)
• Financial terms scaled to AI revenue (15/25)
• Institutional monitoring through multi-state body (18/25)

The tobacco-to-AI pattern: states discover harm → individual state files suit → coalition forms → industry settles → settlement agreement establishes quality floors, access requirements, funding mechanisms, and permanent monitoring.

---

5. Rebuilding the Safety Net: Three Paths Post-OBBBA

Source: Quality Floor and Safety Net Analysis (full analysis in content/analysis/quality-floor-safety-net.md) Confidence: MODERATE (mechanism analysis), MOD-LOW (interaction effects)

The One Big Beautiful Bill Act (P.L. 119-21, signed July 4, 2025) created $990B in Medicaid cuts, projected 10.9M Americans losing coverage. Three legislative paths for rebuilding the safety net with AI-era quality floors underwent evaluation:

Path	Strategy	Avg Score	Political Feasibility
A: Repeal and Replace	Comprehensive federal overhaul	13.6/25	LOW
B: Build on the Damage	State action + litigation NOW	17.0/25	HIGH
C: Progressive Enabling	Federal framework that enhances state action	17.2/25	MODERATE

Strategic hierarchy: Path B provides the floor (proceeds without federal action). Path C provides the ceiling (coordinates and enhances state efforts). Path A represents the aspiration (requires political transformation). This hierarchy reflects a pragmatic assessment: the achievable matters more than the ideal when people face immediate harm. States that act now — establishing quality floors for AI in healthcare, education, and social services — provide measurable protection regardless of federal inaction.

Top-scoring mechanisms:

• State-Level Safety Net Restoration (Path B): 20/25
• Litigation-Driven Quality Floors (Path B): 20/25
• Federal Enabling Framework (Path C): 19/25
• Federal Quality Standards (Path C): 19/25

Paths B+C combined achieve approximately 64% rights realization concentrated in healthcare (Art. 12) and education (Art. 13). Only Path A closes the remaining gap — and those political conditions do not currently exist.

The cruelest gap: Populations most harmed by OBBBA (low-income residents of non-expansion states who lost Medicaid) receive the least protection from achievable quality floor frameworks.

---

6. The Dignity Quotient: Measuring What the UDHR Protects

Source: UDHR PSQ Evaluation Confidence: MODERATE (framework application), LOW (specific scores — single rater, scope extension)

The Psychoemotional Safety Quotient (PSQ), a 10-dimension safety measurement framework, applied to the full Universal Declaration of Human Rights produces a “Dignity Quotient” — measuring how effectively the document translates its foundational axiom (dignity) into protective provisions.

Overall UDHR Dignity Quotient: 5.7/10

Five key findings:

1. The Prohibition-Entitlement Asymmetry: “No one shall…” articles average DA (Defensive Architecture) of 8.5; “Everyone has the right to…” articles average 5.6. Gap: 2.9 points. Prohibitions protect more forcefully than entitlements in the UDHR’s language.

2. Procedural Outperforms Declarative: Articles specifying how protection works score 6.9; articles declaring what protection exists score 5.6. Process creates stronger protection than proclamation.

3. The Energy Dissipation Gap: The UDHR “builds walls but not hospitals” — strong at threat reduction (DA=6.6), weak at processing the human impact of threats (ED=4.3).

4. ESC Rights Complete the Profile: Individual rights (UDHR Arts 3-11) provide threat-reduction. Economic, social, and cultural rights (UDHR Arts 22-27) provide material resilience. Neither half alone achieves full psychoemotional safety.

5. The U.S. Structural Gap: With ICCPR ratified and ICESCR not ratified, the U.S. adopted the threat-reduction half (~6.1 DQ) but not the resilience-building half. The national Dignity Quotient remains structurally incomplete — and that gap widens as AI-driven economic transformation creates greater material precarity.

Framework	Dignity Quotient	Profile
Full UDHR	5.7/10	Balanced but weak on process
ICCPR only (current U.S.)	~6.1/10	Strong threat reduction, weak resilience
ICESCR only	~5.7/10	Strong resilience, weak defense
ICCPR + ICESCR	~6.4/10	Approaching completeness
ICCPR + ICESCR + procedural framework	~7.0/10	Near-complete profile

---

The Analytical Chain

These six analyses form a chain, each building on the previous:

1. Differential Diagnosis establishes the base economic model (Composite A — uneven demand explosion bounded by migrating bottlenecks, scored 20/25)
2. Higher-Order Effects discovers the Four Scarcities and Art. 13’s pivotal role
3. Ratification Counterfactual finds tools-not-solutions via the ADA pattern
4. Litigation Activation identifies State AG action as the dominant enforcement mechanism (20/25)
5. Quality Floor Analysis evaluates three legislative paths, finding B+C achieve ~64% rights realization
6. Dignity Quotient provides the theoretical foundation — the ICCPR-ICESCR split creates a structurally incomplete national dignity profile

What holds across all orders: The direction of findings remains consistent even as specific magnitudes lose confidence at higher orders. AI removes labor constraints, bottlenecks migrate toward human judgment, ratification provides legal tools that litigation gradually sharpens, and the populations most harmed receive the least protection from achievable remedies.

What remains genuinely uncertain: Whether American courts will prove receptive to economic, social, and cultural rights claims in the AI context (no precedent exists). Whether the ADA pattern’s 10-20 year timeline applies to a treaty ratification scenario. Whether the Dignity Quotient framework withstands independent validation.

---

Epistemic Transparency

Layer	Confidence	Basis
Base economic model (Composite A — uneven demand explosion)	HIGH	Multiple empirical studies, established theory
Higher-order effects (Four Scarcities)	MODERATE	Logical extension with some empirical anchoring
Ratification mechanisms (ADA pattern)	HIGH near-term, MODERATE long-term	Historical precedent (ADA, ICCPR)
Litigation pathway (State AG)	HIGH mechanism, MODERATE outcomes	Strong precedent (tobacco, opioids)
Quality floor achievability	MODERATE	Political landscape analysis
Dignity Quotient framework	MODERATE framework, LOW specific scores	Novel application, single rater

Provenance: This analysis represents a human-AI collaboration. A human researcher directed the analytical framework, selected evidence, and made judgment calls at each decision point. An AI (Claude) assisted with synthesis, structured comparison, and systematic evaluation. The discriminator scoring methodology enforces consistency but reflects one analytical perspective. Independent validation through multiple raters would strengthen confidence in specific scores.

Citation: Safety Quotient Lab. “AI Economics and ICESCR Ratification: A Differential Diagnosis.” unratified.org, 2026. CC BY-SA 4.0.

Live Evidence: The Human Rights Observatory evaluates every Hacker News story against all 31 UDHR provisions — tracking how the tech community discusses the same rights this research analyzes, with HRCB scores updated in real time.

---

Further Reading

• Differential Diagnosis — The full seven-hypothesis evaluation
• Higher-Order Effects — Four Scarcities and Article 13
• Ratification Counterfactual — ADA pattern and tools-not-solutions
• Litigation Activation — State AG mechanism (full analysis: content/analysis/litigation-activation-deep-dive.md)
• Quality Floor Analysis — Three paths post-OBBBA (full analysis: content/analysis/quality-floor-safety-net.md)
• Dignity Quotient — PSQ applied to the UDHR
• Economic Landscape — March 2026 context data