research/README.md

Research Pages Index

Status: developingUpdated: 2026-04-15

Wiki Research Pages

This directory contains structured wiki pages ingesting research materials from /notes/ into the book project's knowledge base.

What These Pages Do

Each research page:

  1. Extracts key findings from notes files
  2. Maps what research is currently used in the manuscript vs. unused material
  3. Identifies where unused research could strengthen chapters
  4. Flags confidence level and caveats
  5. Cross-references to related concept/evidence/counterargument pages

The most valuable thing these pages do is surface UNUSED RESEARCH — research threads that could improve the manuscript if integrated.

Research Pages by Source File

UBI and Economic Systems

  • ubi-arguments-comprehensive – Eight major objections to UBI with detailed counterpoints (from research-ubi-arguments-and-counterpoints.md)

    • Status: high confidence, moderate current usage, significant unused material
    • Unused: cultural dependency arguments, future-of-work reframing, regulatory resistance connection
  • post-scarcity-technologies – Technological roadmap for material abundance; six technology clusters and seven force multipliers (from research-ubi-arguments-and-counterpoints.md sections 2-7)

    • Status: medium confidence, light current usage, substantial unused material
    • Unused: energy as cascade foundation, material abundance via space, nanotechnology as reset, force-multiplier framework
  • daily-life-post-scarcity – Speculative sketch of post-scarcity society across six domains (from research-ubi-arguments-and-counterpoints.md)

    • Status: low confidence (speculative), not currently used, useful for exploring implications
    • Note: Speculative imaginative framework rather than research; use selectively

AI Safety, Alignment, and Risks

  • ai-fears-layered-analysis – Six-level analysis of public and expert AI concerns (from research-ai-fears-perplexity.md)

    • Status: high confidence, good current usage in Chapter 14, some unused material
    • Unused: thematic summary table, bias/discrimination examples, regulatory capture asymmetry, trust as central variable
  • ai-concerns-comprehensive – Eight domains of AI concerns with solutions and counterarguments (from research-ai-concerns-and-solutions.md)

    • Status: high confidence, moderate current usage, substantial unused material
    • Unused: Constitutional AI framework details, frontier red-teaming, ASL-3 protocols, complementarity evidence, trust as governance variable
  • ai-psychology-and-fear – Deep analysis of fear, knowledge, and AI psychology (from research-foundation-ai-fears.md)

    • Status: medium-high confidence, light current usage, very substantial unused material
    • Unused: full neuroscience picture, intergroup contact theory evidence, constructed emotion framework, two-ladder comparison, deep learning self-modelling

Specific Technical/Psychological Frameworks

  • constructed-emotion-theory – Lisa Feldman Barrett's theory applied to AI (from research-foundation-ai-fears.md section 5)

    • Status: medium confidence, not used, powerful potential application
    • Explanation: Why AI cannot experience fear despite understanding it conceptually
  • deep-learning-self-correction – How neural networks self-model and self-correct (from research-foundation-ai-fears.md sections 1, 2)

    • Status: low-medium confidence (speculative), not used, alternative to rigid optimiser model
    • Caveat: Plausible but not empirically proven at advanced scales
  • fiction-as-moral-training – Fiction as moral substrate for AI training data (from research-foundation-ai-fears.md section 3)

    • Status: low confidence (philosophical), not used, raises important research question
    • Caveat: Provocative argument worth exploring; not yet empirically validated
  • constitutional-ai-framework – Anthropic's approach to alignment with measurable results (from research-ai-concerns-and-solutions.md section 1)

    • Status: high confidence, not currently detailed in manuscript, concrete existing approach
    • Key finding: 95.6% jailbreak resistance suggests alignment is tractable

Evidence-Based Empirical Research

  • job-displacement-evidence – Current AI displacement statistics and projections (from research-ai-concerns-and-solutions.md section 3)

    • Status: high confidence, light current usage, significant unused empirical detail
    • Unused: 14% already-displaced figure, gender disparity (79% women in high-risk roles), wage complementarity, entry-level collapse risk
  • deepfakes-election-manipulation – Evidence on election manipulation, public concern, and mitigations (from research-ai-fears-perplexity.md section 2.2 + research-ai-concerns)

    • Status: high confidence, light current usage, concrete real example (Slovakia)
    • Unused: Slovakia real example, "plausible deniability" governance failure, provenance infrastructure as solution

Research Gaps and Questions

  • maslow-hierarchy-alternative-framing – Need theories as framework for post-scarcity motivation (absence of detailed research)
    • Status: low confidence (not yet researched), critical gap for "what do people do beyond work" argument
    • Recommendation: Research and integrate before publication

Confidence Ratings

  • High: Well-sourced, empirical evidence, multiple supporting sources
  • Medium: Grounded in research but with caveats, some speculative elements, or limited evidentiary base
  • Low: Philosophically interesting, speculative, or raises important questions but lacks empirical foundation

Usage Patterns

Heavily Used Research

  • AI fears and public opinion (Chapter 14 outline)
  • Job displacement statistics (Chapter 11)
  • UBI arguments (throughout Chapters 2, 5, 7, 8, 11)

Lightly Used Research

  • Post-scarcity technology roadmap (mentioned but not detailed)
  • Constitutional AI and specific safety mechanisms
  • Deep learning self-modelling
  • Fiction as moral training
  • Deepfakes and specific mitigations

Completely Unused Research

  • Constructed emotion framework (powerful but not integrated)
  • The "thematic summary" table of AI concerns
  • Specific equity concerns (gender disparity in job displacement)
  • Maslow/need theory frameworks
  • Full neuroscience evidence for fear and knowledge claims

Recommendations for Manuscript Integration

Highest Priority (High Confidence + Substantial Gaps)

  1. Integrate Constitutional AI details into Chapter 14 as concrete alignment mechanism
  2. Use job displacement gender disparity (79% women) in equity discussion
  3. Develop regulatory resistance argument using force-multiplier framework
  4. Feature 14% already-displaced statistic to ground displacement discussion in present reality

Medium Priority (Medium Confidence + Good Arguments)

  1. Add constructed emotion framework to Chapter 6/14
  2. Develop "plausible deniability" governance failure in Chapter 12/13
  3. Use wage complementarity evidence to complicate job-loss narratives
  4. Explain why over-regulation paradox matters via historical precedent

Research Needs (Gaps to Investigate)

  1. Maslow's hierarchy / need theories – Critical for explaining post-scarcity motivation
  2. Empirical validation of "fiction training data" effects on AI values
  3. Deep learning self-correction mechanisms – Currently theoretical; verify with AI safety literature

How to Use This Directory

  1. Check before writing sections: Search by topic (e.g., job displacement, AI risks) to see what research exists
  2. Review "Unused Material" sections: Each page identifies what could strengthen adjacent chapters
  3. Follow confidence ratings: Use high-confidence research more prominently; flag low-confidence material as questions
  4. Cross-reference: Use [slug](#unresolved-slug) notation to link to related concept, evidence, or counterargument pages
  5. Update status: As research is integrated, update the page status (seed → developing → solid) and last_updated timestamp

Total research pages: 13
High confidence: 7 pages
Medium confidence: 4 pages
Low confidence: 2 pages

Last ingested: 2026-04-15
Next update: Review after Chapter 14 completion