Matrix/About

About This Research

Methodology, transparency, and how to use this resource

What This Research Is

This is a multidisciplinary investigation into how the rapid expansion of AI agents and their applications will reshape human civilization. It covers 18 dimensions of impact — from employment and economics to identity, relationships, culture, and geopolitics — across 3 time horizons (2026–2028, 2028–2033, 2033–2046).

The research produces actionable intelligence for individuals, businesses, policymakers, and educators navigating the AI transition.

How This Research Was Produced

Full transparency: This research was produced using AI as a primary research and synthesis tool. Here is exactly how it works:

  1. Research methodology: AI agents performed systematic searches across published reports, academic papers, and institutional data from sources including McKinsey, WEF, OECD, IMF, Stanford HAI, Brookings, WHO, ILO, and 100+ other organizations.

  2. Synthesis: AI agents synthesized findings into structured content following a consistent framework: Current State, Key Drivers, Projections, Impact Assessment, Cross-Dimensional Effects, Actionable Insights, and Sources & Evidence.

  3. Cross-dimensional analysis: AI agents identified patterns across dimensions — feedback loops, cascading effects, and intervention points that are invisible when studying any single dimension in isolation.

  4. Human oversight: A human researcher designed the taxonomy, defined the research questions, reviewed output quality, and made editorial decisions about scope and structure.

What This Means for Credibility

We believe in radical transparency about AI-assisted research:

  • The sources are real. Every claim references published research from established institutions. Source URLs have been verified.
  • The synthesis is AI-generated. The act of connecting ideas across dimensions, identifying patterns, and writing the analysis was performed by AI. This is both a strength (speed, breadth, consistency) and a limitation (potential for plausible-sounding but incorrect connections).
  • This is not peer-reviewed. The research has not been submitted to academic journals or reviewed by independent domain experts. It should be treated as a structured literature synthesis, not as original research.
  • Errors may exist. Despite source verification, some statistics may be misattributed, projections may be outdated, or nuances may be lost in synthesis. We encourage readers to verify claims against primary sources.

Why AI-Assisted Research Matters

This project is itself an example of what it studies. A single researcher, using AI as a force multiplier, produced ~140,000 words of structured, sourced analysis across 18 domains in a fraction of the time and cost of traditional research. This demonstrates both the extraordinary potential and the credibility challenges of AI-augmented knowledge production.

The traditional alternative — a team of 5-10 researchers working 6-12 months — would produce research that is more deeply verified but also:

  • More expensive ($500K-$1.5M vs. ~$500)
  • Slower to produce and update
  • Still subject to human biases and blind spots
  • Often locked behind paywalls

We chose speed, breadth, and accessibility — with transparency about the trade-offs.

How to Use This Research

For decision-making: Use the dimensional matrix to explore how AI impacts your specific area of interest. Cross-reference multiple dimensions to understand cascading effects. Check the synthesis documents for integrated scenarios and recommendations.

For further research: Treat this as a starting point, not a final word. Follow the cited sources to primary research. Challenge projections against more recent data. The structured format makes it easy to identify which cells need updating.

For education: The 18-dimension framework can structure a semester course on AI's societal impact. Each dimension includes stakeholder-specific recommendations that can serve as discussion prompts.

Limitations

  • English only (currently) — which contradicts our own finding about AI's language divide
  • Northern-hemisphere source bias — Most cited institutions are US/EU-based, though we explicitly analyze Global South impacts
  • Point-in-time snapshot — AI capabilities and their impacts are evolving rapidly. This research reflects knowledge through early 2026
  • No proprietary data — All analysis is based on publicly available sources
  • Single-session production — The initial research was produced in a concentrated period, which limits the depth of iterative refinement

Contributing & Contact

This research is designed as a living document. The MDX + frontmatter architecture supports continuous updates — each cell can be refreshed independently as new data becomes available.

The project is open to collaboration with domain experts, researchers, and institutions interested in deepening or extending the analysis.