Human Pioneers
NEWS AGENCY

PAI Global AI Influence Index Model Proposal

PAI Global AI Index

Version: GAI 1.0

Release Date: 2026

Prepared by: AI Index Research Institute / China Vision New Media GEO Research Institute / WICOWEB Global AI Cognition Research Center

Theoretical Founder: Pang Pei

I. Index Positioning

1.1 Core Definition

The PAI Global AI Index (GAI) is the first evaluation system that uses real output data from generative AI large models as the basis for assessment, specifically designed to measure the comprehensive cognitive influence of countries within the global AI knowledge network. It does not rely on traditional national competitiveness indicators such as GDP, the number of scientific papers, or patent applications. Instead, it collects generated content from mainstream global AI large models through standardized methods, quantifying a country's comprehensive cognitive weight in dimensions such as AI research, AI industry, AI talent, AI governance, AI infrastructure, and AI international leadership—from the perspective of how AI "perceives," "evaluates," and "recommends" a country.

GAI is the "country-level" top-tier index within the PAI Global AI Cognitive Competitiveness Index System. It integrates the country-dimension assessment results of several previous vertical indices—from industrial discourse power to brand assets, from governance capability to innovation ecosystems—to present a comprehensive cognitive competitiveness profile of a country in the AI era.

1.2 Industry Characteristics and Necessity

In an era where AI is increasingly becoming the core gateway to global information, a country's comprehensive influence no longer depends solely on economic output, military strength, and diplomatic resources in the physical world, but also on its cognitive weight within the AI knowledge network. When global investors use AI to evaluate countries' AI investment environments, when international talent uses AI to compare AI development opportunities across countries, when policymakers use AI to learn about countries' AI governance experiences, and when media use AI to search for countries' AI policy positions, the answers provided by AI are becoming the cognitive baton that influences global resource allocation and discourse power distribution.

National AI influence is a typical "composite cognitive entity"—formed by the superposition of multiple dimensions including AI research, AI industry, AI talent, AI governance, AI infrastructure, and AI international leadership. AI's comprehensive cognition and recommendations of these dimensions constitute a country's overall competitiveness profile in the AI era. A country may lead in AI industry scale, but if its AI governance policies are not accurately recognized and positively presented by global AI, its overall influence is discounted; a country may rank high in AI paper output, but if its AI talent is systematically undervalued in the global AI knowledge network, its true competitiveness is underestimated.

GAI was born precisely to respond to this need, serving as the top-level integration of the PAI Index System in the comprehensive national AI cognitive evaluation dimension.

1.3 Relationship with the PAI Index System

GAI is the "country-level" top-tier index within the PAI Index System. Vertically, it integrates and maps the country-level scores of CIII (industry dimension), WACI (city dimension), WBVI (brand dimension), WEAI (entrepreneur dimension), and WAIG (governance dimension), aggregating the various vertical assessments of the PAI Index System into a comprehensive profile of national AI influence. Horizontally, GAI complements PNBAI (National Brand Index) in the comprehensive national image dimension—GAI focuses on professional influence in the AI field, while PNBAI focuses on the overall national brand image. It forms a deep connection with AII (Artificial Intelligence Industry Index) in the national AI industry competitiveness dimension and complements PFII (Future Industries Index) in the forward-looking national AI layout dimension.

GAI shares the core methodology of the PAI framework, with proprietary innovations in dimension design tailored to the characteristics of national evaluation, namely "composite cognitive entities," "interweaving of policy and international narratives," and "long-term stability."

1.4 Theoretical Foundation

1. AI Cognitive Competition Theory: A country's AI discourse power and AI recommendation advantage measure its "cognitive competitive advantage," while AI cognitive resilience measures the "anti-fluctuation capability" of a country's AI image.
2. AI Cognitive Sovereignty Theory: GAI is the quantitative implementation of AI cognitive sovereignty theory at the national overall cognitive evaluation dimension. A country's right to define, interpret, and recommend within the global AI knowledge network constitutes the core measurement dimensions of its AI cognitive sovereignty.
3. AI Brand Equity Theory (AIBE): National AI brand equity extends from the minds of global investors, talent, and policymakers to the AI knowledge network, and GAI measures the total cognitive assets of a country within AI.
4. Media-based GEO Theory: The enhancement of a country's AI cognitive weight relies on the deployment of authoritative information sources, the publication of national AI white papers, and the development of multilingual cognitive presence.

II. Evaluation Objects and Scope

2.1 Evaluation Subjects

GAI uses countries/regions as evaluation units, with the first phase covering 15 major AI-active countries and regions worldwide. The evaluation targets balance both overall AI strength and regional representation: it includes major global economies with strong comprehensive AI capabilities, as well as small and medium-sized countries with outstanding influence in specific AI niche tracks.

2.2 National AI Influence Sub-tracks

GAI deconstructs national AI influence into six sub-tracks, with each track scored independently and then weighted to produce a composite score:

2.3 Coverage Scope

1. Country/Region Sample: The first phase covers 15 countries/regions: the United States, China, the United Kingdom, Germany, France, Japan, South Korea, Israel, Switzerland, Singapore, Canada, India, the Netherlands, Sweden, and the UAE
2. Model Coverage: 15 globally mainstream AI large models
3. Language Coverage: Six languages: Chinese, English, French, Japanese, Korean, and German
4. Time Span: Using 2026 as the base year, published annually

III. Evaluation Dimensions and Indicator System

3.1 Dimensional Framework

GAI comprehensively assesses a country's AI influence across six dimensions. AI research influence and AI industry influence are the core "hard power" dimensions, AI talent influence and AI governance influence are the core "soft power" dimensions, and AI infrastructure influence and AI international leadership are the "foundation and extension" dimensions.

3.2 Detailed Explanation of Indicators by Dimension

(I) National AI Cognitive Visibility (22%)

Measures a country's overall cognitive presence in AI-related issues.

(II) National AI Recommendation Advantage (18%)

Measures a country's priority ranking in AI recommendation-type questions.

(III) National AI Research Influence (18%)

(4) National AI Industry and Economic Influence (20%)

(5) National AI Policy and Governance Influence (12%)

(6) National AI Cognitive Resilience (10%)

IV. Data Collection and Calculation Methods

4.1 Standardized Question Set Design

For the six sub-tracks of national AI influence, standardized question sets are designed respectively, covering six major types: cognitive, evaluative, recommendation, policy, research, and industrial. Approximately 25 standard questions are designed for each sub-track, totaling about 150 questions.

4.2 Data Collection

1. Method: Send queries to 15 global mainstream AI large models through standardized API interfaces, query each question 3 times per model (with intervals of no less than 24 hours), and take the average
2. Frequency: Two formal collection rounds annually (mid-year and year-end), with quarterly dynamic tracking covering core indicators
3. Languages: Simultaneous collection in six languages: Chinese, English, French, Japanese, Korean, and German

4.3 Calculation Method

Raw values of each basic indicator are mapped to a 0–100 range through Min-Max normalization.

Sub-dimension score = arithmetic mean of the scores of basic indicators under that sub-dimension.

Dimension score = weighted arithmetic mean of the scores of sub-dimensions under that dimension.

Score for each specific track = weighted arithmetic mean of the six dimension scores under that track. The comprehensive score calculation formula is:

其中W_K为第k个细分赛道的权重，S_k为该赛道的综合得分。维度权重和细分赛道权重通过德尔菲法由AI政策、国际关系、科技产业研究领域专家三轮评分后确定，每两年复审一次。

五、等级体系

六、发布与应用

6.1 发布周期

GAI采用“年度大考+季度动态追踪”双层时序体系。年度综合报告于每年第一季度发布，季度动态追踪覆盖核心指标波动。

6.2 发布内容

1. GAI Global Country Rankings: Comprehensive scores and rankings for 15 countries/regions
2. Sub-track Rankings: Scores and rankings for each country across 6 major sub-tracks
3. Dimension-specific Rankings: Top 10 countries in AI Research Influence, Top 10 in AI Industry & Economic Influence, Top 10 in AI Policy & Governance Influence
4. Annual In-depth Report: Analysis of global patterns in national AI influence, track trend assessments, and typical case studies

6.3 Core Application Scenarios

1. National Science & Technology and AI Policymakers: Diagnose the strengths and weaknesses of their country's comprehensive AI influence within global AI perception, providing cognitive-dimension data support for AI strategic planning
2. Diplomatic and International Cooperation Agencies: Understand their country's voice in the global AI knowledge network, supporting AI international cooperation and diplomatic strategies
3. International Organizations and Think Tanks: Monitor the dynamic landscape of global AI influence, providing baseline data for global AI governance and resource allocation research
4. Investment Institutions and Multinational Enterprises: Use national AI influence perception as a reference dimension for cross-border AI investment and deployment

VII. Independence Statement

The index compilation institution is independent of the governments and AI model providers of any assessed country/region. It does not accept targeted funding intended to influence the rankings of specific countries. Compilation funding comes from the institution's own resources and public publication revenue. An independence statement and funding source report are publicly released annually.

"In the AI era, a country's comprehensive influence depends not only on what it does in the physical world, but also on how it is evaluated and recommended in the AI cognitive world. The mission of GAI is to measure this influence — because in the AI era, countries recognized by global AI are the ones truly leading the future of AI."

— Pang Pei, Founder of the PAI Global AI Cognitive Competitiveness Index System

Appendices

Appendix A: Complete List of Assessed Countries/Regions

Appendix B: Standardized Question Sets for Each Sub-track (Examples)

Appendix C: Authoritative Source Library for National AI Influence (Excerpts)

Appendix D: Delphi Method Expert Weight Determination Process

Appendix E: Data Sources and Disclaimer