Research on the Global paradigm, Industrial      Restructuring      and Grassroots  Implementation   path of In-depth Integration of Al and Digital Intelligence

Authors:  Fnu Oudom

Abstract

Accelerated iterations of general artificial intelligence technologies worldwide

have made the integration of digital intelligence the core thread of the Fourth

Industrial Revolution and the underlying logic for fostering new productive forces and reshaping the global competitive landscape of the digital economy. Based on a comparison of national AI strategies across multiple global economies, this

paper constructs a three-dimensional analytical framework covering global

landscape, top-tier industries and grassroots implementation. It systematically

interprets the theoretical origin and internationally differentiated development

paradigms of AI-digital intelligence integration. Covering five core sectors

including advanced manufacturing, digital finance, smart agriculture, urban

governance and digital trade, this paper extracts high-end transformation

highlights in each field. It further extends its research to grassroots scenarios such as towns, communities, micro, small and medium-sized enterprises (MSMEs), and rural governance, analyzing bottlenecks, core starting points and practical key

points for the downward penetration of digital intelligence. This study identifies

common challenges arising from global integration, including technological

divides, algorithm governance, circulation of data factors and regional imbalance. Coordinating the logic of international collaboration and hierarchical domestic

advancement, it proposes a four-in-one long-term development path featuring

technological independence, industrial ecology, grassroots inclusiveness and

global digital governance. Boasting an international vision, industrial depth and

practical grassroots value, this research provides theoretical support and practical references for the construction of Digital China and the coordinated development of the global digital economy.

keywords: Integration of Digital Intelligence; Generative AI; New Productive Forces; Global Digital Competition; Grassroots Digital Governance; Industrial Digitalization; Data Factors

1 Introduction

1 . 1 Research Background and contemporary value

Since the industrialized application of generative large models, AI has transitioned from a single-point tool to a universal technology applicable across all fields,

propelling the digital economy into a new stage driven by AI-enabled digital

intelligence integration. Globally, major economies including the United States, the European Union, China, Japan, South Korea and emerging economies have

successively rolled out national-level artificial intelligence strategies, regarding the integration of digital intelligence as a core lever to seize industrial

commanding heights and reshape the international division of labor system.

Domestically, the Overal Layout Plan for Digital China Construction and the New Generation Artificial Inteligence Development Plan explicitly propose empowering the real economy with artificial intelligence and advancing digital inclusiveness in urban and rural areas. The core essence of developing new productive forces lies  in leveraging the deep integration of AI, big data and computing power networks to break reliance on traditional production factors and achieve systematic leaps in total factor productivity. Currently, the integration of digital intelligence presents a polarized pattern: first-tier cities and large industrial chain enterprises have

realized full-process intelligent transformation, while county-level regions, rural areas, micro and small business entities and grassroots government units

generally suffer from insufficient downward penetration of technologies,

superficial application and unbalanced input-output ratios. The contradiction of "surplus top-tier technologies yet inadequate grassroots applications" has

become a core constraint hindering in-depth integrated development.

Internationally, three differentiated development paths for global digital

intelligence integration have taken shape: market-driven free innovation,

regulation-restrained development, and government-coordinated advancement. Developed economies have built technological barriers relying on their

advantages in computing power, models and data, while developing countries

face multiple constraints such as digital divides, high computing costs and talent shortages, exacerbating imbalanced global digital development. There is an

urgent need to build an inclusive and universally beneficial global governance system for digital intelligence integration.

Against this backdrop, establishing a complete research chain of "global

competition landscape – high-end industrial transformation – inclusive grassroots implementation" and systematically analyzing the internal mechanisms of AI-

digital intelligence integration, industrial innovation highlights, key grassroots implementation points and global collaborative paths carries significant

theoretical innovation value and practical significance.

1 . 2 Domestic and Foreign Research Review

1 .2 . 1 Foreign Research context

Academic circles in Europe and the United States focus on two major research

directions. First, from a techno-economic perspective, scholars demonstrate the

reconstructive effects of AI as a general technology on entire industrial chains,

taking industrial practices of Google and OpenAI as samples to measure the

transformative impact of large models on production efficiency and business

models. Second, from a digital governance perspective, centered on the EU’s AI

Act, studies focus on algorithm ethics, cross-border data flows and hierarchical

risk supervision frameworks for AI, alerting to risks of technological monopoly and digital exclusion. Special reports released by the United Nations Conference on

Trade and Development (UNCTAD) point out that developed economies achieve

far higher efficiency in industrialized AI implementation than developing

economies, and MSMEs worldwide face three bottlenecks in digital transformation:

insufficient computing power, limited data access and inadequate technical skills, making inadequate technological inclusiveness a universal global challenge.

Foreign research has notable limitations: rooted in Western market governance systems, it lacks analysis of localized scenarios including the urban-rural dual

structure, grid-based grassroots governance and whole-of-nation system research breakthroughs, making it incompatible with China’s integrated

development model featuring "top-down top-level design plus bottom-up grassroots practice".

1 .2 . 2 Domestic Research status

Existing domestic research outcomes fall into three major categories. First,

theoretical research on the integration of new productive forces and AI, clarifying the underlying logic through which digital intelligence integration reshapes

production factors and industrial structures. Second, case analysis of regional and industrial digital transformation, focusing on industrial internet and smart city

construction. Third, single-scenario research on digital villages and grassroots government administration.

Obvious gaps remain in current research: there is a lack of integrated research

combining horizontal global comparison, comprehensive industrial review and

systematic grassroots implementation. Most studies are confined to a single

industry or region, failing to connect the logic of global competition, high-end

industrial upgrading and grassroots digital inclusiveness. Systematic extraction of practical points and transformation pain points for grassroots entities such as

counties, villages and micro and small enterprises is also absent.

1 . 3 Research ldeas, Framework and Innovations

1 .3 . 1  Research ldeas

This paper follows the main research line: theoretical mechanism – global

landscape – top-tier industrial highlights – key grassroots implementation points

– practical challenges – collaborative domestic and international development

paths. Taking international digital competition as the macro perspective, covering high-end transformation across all industries at the meso level, and delving into grassroots governance and market entities at the micro level, it achieves a trinity of "global strategic vision, industrial depth and grassroots practical precision".

1 .3 . 2 core Innovations

1 .    Vision Innovation: Construct a comparative framework for three major global AI development paradigms, distinguishing digital intelligence integration paths of developed countries, the EU and emerging economies to break the limitation of  single domestic research perspectives.

2.    Dimensional Innovation: Connect the dual layers of "high-end industrial upgrading and inclusive grassroots implementation". It extracts benchmark industrial models while systematically sorting out digital intelligent practical guidelines for towns, communities, villages and micro and small enterprises.

3.    Practical Innovation: Moving beyond pure technical analysis, it integrates four dimensions of technology, institutions, factors and governance, proposing

systematic solutions balancing international competition security and balanced grassroots inclusiveness.

2 core Theoretical connotation and Internal operating Mechanisms ofAl-Digital Intelligence Integration

2. 1 Definition of Digital Intelligence Integration

Digital intelligence integration is not a simple superposition of digital technologies and artificial intelligence. Taking data as the core production factor, it relies on

digital infrastructure including computing power networks, large models, the

Internet of Things (IoT), digital twins and blockchain to form a new economic and social development model featuring two-way empowerment of "digital

infrastructure + intelligent brain". Digitalization completes universal

interconnection and full-scale data collection, while artificial intelligence realizes

data value mining, autonomous decision-making and self-adaptive optimization

for scenarios. The deep coupling of the two reconstructs the entire chain of

production, distribution, circulation and governance, serving as the core carrier for the digital economy to advance into the stage of new productive forces.

Following the popularization of generative AI, digital intelligence integration has

formed three new characteristics: diversified factors (new elements including word tokens, models and intelligent agents), naturalized interaction (multimodal human -machine collaboration), and universalized application (lightweight models

lowering the threshold for grassroots use).

2. 2Three-Dimensional Empowerment operating

Mechanisms

2.2 . 1 Factor Restructuring Mechanism: Reshaping the Three Traditional Factors of production

Laborers: AI intelligent tools liberate workers from repetitive labor, pushing

laborers to transition toward high-skill creative work, algorithm operation and

maintenance, and digital intelligence management, forming new human-machine collaborative labor subjects.

Means of labor: Traditional equipment, production lines and government

administration tools are upgraded into intelligent terminals and digital twin systems with autonomous perception and self-optimization capabilities.

Objects of labor: Data becomes an independently tradable asset, and massive

unstructured data from industries, government administration and people’s

livelihoods is transformed into value-adding production materials, expanding the boundaries of economic activities.

2.2 . 2 Industrial upgrading Mechanism: Three-stage terative Transformation

Basic Layer: Equipment networking and data connectivity to resolve data silos.

Intelligent Layer: Embedding industry-specific large models into business processes to realize prediction, optimization and autonomous scheduling.

Ecological Layer: Interconnected data across the entire industrial chain to build a symbiotic digital intelligence industrial ecosystem, driving industries toward high- end, intelligent and green development.

2.2 .3 Governance Inclusiveness Mechanism: Flat and precise public services

AI breaks data barriers within hierarchical administrative systems, enabling cross- departmental sharing of government data through unified digital platforms.

Algorithms match differentiated demands of residents and enterprises, shifting grassroots governance from "empirical extensive management" to "data-driven precise governance" and narrowing gaps in public services between urban and rural areas and across regions.

3comparative Analysis of Differentiated Global

Development paradigms forAl-Digital Intelligence Integration

Three mature development models for global digital intelligence integration have taken shape. Each country has adopted distinct advancement paths based on its resource endowments and institutional systems, featuring coexistence of

competition and cooperation, offering international references for China’s

hierarchical advancement of digital intelligence integration.

3. 1 The united states: Market-Driven , Technology- prioritized Free Innovation paradigm

1 .    Top-level Layout: Relying on the CHIPS and Science Act and the "Stargate"

supercomputing infrastructure initiative, hundreds of billions of capital has been  invested in general large models and autonomous intelligent agents. The Defense Advanced Research Projects Agency (DARPA) coordinates cutting-edge basic AI  research, with tech giants fully leading industrial implementation.

2.    Core Advantages: Leading globally in computing chips, underlying large

models and open-source ecosystems, with rapid commercialization. Its digital

intelligence integration in finance, high-end manufacturing and the internet sets global benchmarks.

3.    Inherent Shortcomings: Lagging regulation, prominent algorithmic

discrimination, data monopolies and the digital wealth divide. Insufficient

investment in digital intelligence for grassroots public services leads to significant urban-rural gaps in digital inclusiveness.

4.    Underlying Logic: Technological iteration driven by market capital, prioritizing commercial industrial returns while weakening balanced public inclusiveness.

3. 2The European union: Regulation-First, Human- centered Trustworthy Digital Intelligence paradigm

1 .    Top-level Layout: The world’s first AI Act establishes a hierarchical risk

supervision system imposing strict constraints on high-risk AI applications. The Horizon Europe Fund supports lightweight AI transformation for MSMEs and

builds a unified European digital market.

2.    Core Advantages: Strong discourse power in global AI governance rule-

making, sound systems for algorithm ethics and privacy protection, and mature digital transformation supporting services for manufacturing MSMEs.

3.    Inherent Shortcomings: Insufficient self-supply of computing power and

underlying models, slower industrial integration innovation compared with China and the US, and unbalanced digital infrastructure development across member  states.

4.    Underlying Logic: Defining technological boundaries through legal systems, steadily advancing industrial digital transformation under the premise of safety and compliance, while balancing workers’ rights and public value.

3.3 china and East Asian Economies: Government-

coordinated paradigm Balancing Industry and Inclusiveness

Represented by China, South Korea and Singapore, this model features a

distinctive path of "national strategic guidance + market industrial implementation + grassroots inclusive safeguards".

1 .    China: Relying on special projects for new-generation artificial intelligence, a national integrated computing power network and market-oriented reform of

data factors, China coordinates breakthroughs in high-end industries alongside the construction of digital villages and grassroots smart government services,  balancing self-reliance in science and technology and common prosperity.

2.    South Korea / Singapore: National-level AI industrial funds are established to prioritize smart manufacturing and smart cities, while simultaneously advancing  lightweight digital services in counties and communities.

3.    Model Advantages: Concentrated resources to fill gaps in computing power and infrastructure, balanced development gaps between urban and rural areas, large enterprises and MSMEs, and complete supporting policies for downward penetration of digital intelligence at the grassroots level.

4.    Competitive Shortcomings: Core bottlenecks persist in underlying chips and high-end industrial large models, and discourse power in formulating global

digital standards needs to be enhanced.

3. 4common Global Development Dilemmas

1 .    Digital Divide: Widening gaps in computing power, data and talent between developed and developing economies.

2.    Governance Divergence: Inconsistent national AI supervision rules and high

barriers to cross-border data flows, lacking unified global coordination standards that hinder digital trade circulation.

3.    Structural Imbalance: Global digital intelligence resources are concentrated in leading enterprises and core cities, while integration implementation lags in

counties, rural areas and micro and small enterprises.

4 High-End Transformation Highlights and ToP-Tier Application system of Digital Intelligence Integration Across Five core Industries

Based on the commanding heights of global industrial competition, this section

covers five pillar sectors of the real economy: manufacturing, finance, agriculture, urban governance and digital trade. It extracts high-end innovative models and benchmark highlights for each field and constructs a top-tier implementation

framework for industrial digitalization.

4. 1 Advanced Manufacturing: Full Industrial chain

Intelligence Driven by Industrial Large Models (core

Industrial Track)

High-End Industrial Highlights

1 .    Full-Lifecycle Intelligent Manufacturing via Digital Twins: Construct three-

dimensional digital mirrors of factories, production lines and products. AI

simulates production processes in real time, boosting efficiency in fault prediction and capacity optimization by over four times and drastically cutting trial

production costs.

2.    Deployment of Vertical Industry-Specific Large Models: Develop exclusive

industrial large models for automobiles, equipment manufacturing and chemical   industries, connecting full-chain data covering R&D, manufacturing, warehousing, logistics and after-sales maintenance to realize flexible customized production.

3.    Leading Enterprises Driving the Digital Intelligence Ecosystem of Industrial Clusters: Leading enterprises open industrial data platforms to enable upstream and downstream supporting MSMEs to access intelligent systems at low cost,

resolving capital bottlenecks for small business transformation.

4.    Integration of Digital Intelligence and Green Development: AI energy

consumption optimization algorithms dynamically adjust energy use of production equipment, simultaneously realizing smart manufacturing and carbon peaking and carbon neutrality goals, creating benchmark zero-carbon smart factories.

ToP-Tier Implementation Logic: Four-layer architecture of

computing power base - industrial internet platform - industry large model - workshop intelligent terminals.

4. 2 Digital Finance: Trustworthy AI Restructuring cross- Border and Inclusive Financial systems

High-End Industrial Highlights

1 .    Central Bank Digital Currency (CBDC) + AI Cross-Border Settlement: Intelligent algorithms automatically match exchange rates and risk control reviews for cross- border transactions, drastically shortening capital turnover cycles for international trade and building an independently controllable cross-border digital financial

channel.

2.    Grassroots Inclusive Intelligent Risk Control: Lightweight AI models deployed in county rural commercial banks and village banks realize unsecured credit

evaluation based on operational data of farmers and MSMEs, alleviating financing difficulties at the grassroots level.

3.    Generative AI Robo-Advisors: Multimodal AI provides customized asset

allocation solutions for retail investors, lowering the threshold for professional financial services.

4.    Collaborative Risk Prevention via Blockchain and AI: Real-time identification of illegal fund-raising and credit fraud to build a full-domain financial risk early

warning system.

4.3smart Agriculture: Digital Intelligence Integration Empowering Modern Agriculture and Rural Industrial Revitalization

High-End Industrial Highlights

1 .    Full-Scale Intelligent Perception System for Farmland: Satellite remote sensing and IoT terminals collect data on soil, meteorology and crops. AI automatically

generates schemes for irrigation, fertilization and pest control to achieve large- scale cost reduction and efficiency improvement.

2.    Full-Link Digital Traceability for Agricultural Products: AI visual recognition

combined with blockchain realizes full traceability from farmland to dining tables, building regional branded digital agricultural products.

3.    Integrated Rural Digital Industries: AI intelligent live-stream operation and agricultural product market price forecasting connect online and offline

production and sales channels to activate distinctive rural resources.

4.    Digital Management for Smart Breeding: AI vision monitors livestock and

poultry health status and automatically adjusts breeding environments to reduce epidemic risks.

4. 4smart cities and Digital Governance: Full-Domain

Digital Intelligence Brains Building a Modern Governance

system

High-End Industrial Highlights

1 .    Urban Digital Twin Brain: Integrates full-domain data of transportation,

emergency response, people’s livelihoods and government administration. AI

predicts urban congestion, natural disasters and public security risks in advance and realizes integrated scheduling and disposal.

2.    Unified Intelligent Government Service Platform: Interconnected cross-

departmental data enables AI to automatically divert public demands, achieving second-level rapid response for "receipt and immediate handling of public

appeals".

3.    Low-Carbon Smart Cities: AI optimizes energy allocation for power grids, water supply and transportation to create benchmark green intelligent cities.

4. 5 Digital Trade: Al Driving Restructuring of Global Digital

value chains

High-End Industrial Highlights

1 .    Multimodal AI Cross-Border Trade Services: Automatic multilingual translation, overseas market demand forecasting and intelligent customs declaration lower

the threshold for MSMEs to engage in overseas trade.

2.    Trustworthy Cross-Border Data Circulation Platforms: Leveraging data

customs mechanisms to balance data security and shared trade data, building regional digital trade hubs.

3.    Global Market Intelligent Analysis Large Models: Real-time capture of global   industrial, tariff and consumption data to provide decision support for enterprises, overseas layout.

5 Downward penetration ofAl-Digital Intelligence  Integration atthe Grassroots Level: core scenarios, Implementation priorities and key practical points

High-end industrial upgrading constitutes the "superstructure" of digital

intelligence integration, while inclusive grassroots implementation serves as the

foundational support for realizing its value. Focusing on four bottom-tier

scenarios including rural towns, urban community governance, MSMEs and county -level public services, this paper analyzes bottlenecks, core starting points and

practical guidelines for technology penetration, filling the research gap of insufficient grassroots perspectives in existing literature.

5 . 1 scenario 1: Rural Grassroots (Digital villages, Rural Governance, Agricultural Business Entities)

core pain points

1 .    Weak Infrastructure: Insufficient coverage of 5G and computing terminals in remote rural areas, with high costs for data collection.

2.    Low Digital Literacy: Farmers and village committee cadres lack operational capabilities for AI tools.

3.    Superficial Application: Most rural digital practices are limited to short video live streaming, lacking in-depth intelligent applications for production and

governance.

4.    Insufficient Capital Input: Village collectives cannot afford the transformation costs of large models and intelligent equipment.

key Grassroots Implementation Guidelines

1 .    Lightweight Infrastructure Layout: Prioritize deployment of low-cost IoT

perception terminals and lightweight edge AI devices, sharing computing power via county-level computing centers to avoid redundant construction.

2.    Build village-level digital intelligence governance mini-programs integrating village affairs disclosure, dispute mediation, convenient administrative services

and disaster early warning, with AI automatically aggregating villagers’ demands.

3.    Layered Digital Skills Training: Deliver digital intelligence management

training for village cadres and practical courses on AI planting and e-commerce operation for farmers.

4.    Cost Reduction via Government-Enterprise-Village Collaboration: Local

governments and leading agricultural enterprises jointly bear equipment costs, providing free access to lightweight agricultural industry models.

5.    Precise Deployment of Characteristic Scenarios: Prioritize intelligent

traceability and production-sales forecasting for local featured agricultural products to fit actual rural industrial conditions, eliminating vanity projects focused on hardware with negligible practical application.

5 . 2scenario 2: urban community Grassroots (Grid-Based Governance, community Livelihood services)

core pain points

Fragmented data across community departments; insufficient funds for intelligent transformation of old residential compounds; digital exclusion among elderly

groups with high operation thresholds for intelligent devices.

key Grassroots Implementation Guidelines

1 .    Unified community digital intelligence platform: Interconnect data across

multiple departments, enabling AI to automatically identify elderly residents living alone, vulnerable groups and potential safety hazards to deliver proactive services.

2.    Age-friendly lightweight intelligent applications: Simplify operation interfaces and deploy voice-interactive AI services to lower usage barriers for the elderly.

3.    Phased intelligent transformation of old residential compounds: Prioritize deployment of intelligent access control, AI fire hazard early warning and

intelligent garbage sorting supervision, with gradual full-scale promotion.

4.    AI-assisted grid patrols: Cameras combined with computer vision AI

automatically detect illegal street occupation and fire risks, reducing offline workloads for grid administrators.

5.3scenario 3: county-Level MSMEs (Manufacturing, commerce and service Industries)

core pain points

Limited capital to purchase customized industry-specific large models; lack of

professional digital intelligence operation and maintenance talent; vague digital transformation paths with unclear input-output benefits.

key Grassroots Implementation Guidelines

1 .    Government-built inclusive AI service platforms: Unified procurement of lightweight general models with ultra-low service fees for MSMEs to share computing power costs.

2.    Standardized industrial transformation packages tailored for retail, processing and logistics industries to avoid blind investment by enterprises.

3.    Local school-enterprise talent support: Vocational colleges provide part-time AI operation and maintenance services for regional MSMEs.

4.    Prioritize rigid-demand core scenarios: Deploy intelligent inventory

management, AI customer marketing and visual production quality inspection first to rapidly realize cost reduction and efficiency gains.

5 . 4scenario 4: county-Level Grassroots Government Administration (Town convenience service centers,

Grassroots Law Enforcement, Livelihood security) core pain points

Limited staffing and heavy workloads at grassroots institutions; repeated

submission of materials across departments; lagging risk early warning dominated by post-incident disposal.

key Grassroots Implementation Guidelines

1 .    AI automatic government approval: Intelligent verification of materials for

license processing and subsidy applications to reduce pressure on manual service windows.

2.    Full-domain intelligent risk early warning: AI analysis of social insurance, workplace safety and petition data to realize pre-intervention.

3.    Unified aggregation of grassroots data: Connect data across all town offices and stations to eliminate data silos and reduce repeated reporting forms for

grassroots staff.

5 . 5 universal core principles for Grassroots Digital Intelligence Integration

1 .    Inclusiveness First: Reject high-investment, low-practicality vanity high-end

projects, centering on low-cost lightweight technologies.

2.    Demand-Supply Matching: Adapt to real grassroots production and

governance demands, avoiding mechanical replication of large-scale urban industrial intelligent solutions.

3.    Phased Advancement: Fully deploy digital intelligence systems in developed towns while implementing step-by-step construction in underdeveloped regions, prioritizing rigid demands before upgrades.

4.    Digital Inclusiveness: Address gaps in digital capabilities among the elderly,  rural populations and MSME operators, supporting training and simplified offline auxiliary services.

6 prominent challenges Facing Global and Domestic Al -Digital Intelligence Integration

6. 1 International competition and Governance challenges

1 .    Technological Hegemony and Computing Power Barriers: The US and Western countries construct technical barriers through export restrictions on chips and

underlying models, increasing pressure on developing economies to achieve independent and controllable digital intelligence integration.

2.    Fragmented Global Rules: Divergent national AI supervision and cross-border data laws lack unified international coordination standards, hindering digital trade circulation.

3.    Widening Global Digital Divide: Low-income countries lack sufficient computing power, data and talent to share the dividends of AI industries.

6. 2structural challenges in Domestic High-End Industrial Transformation

1 .    Shortcomings in Independent Underlying Technologies: High reliance on  foreign suppliers for high-end industrial large models, AI computing chips and professional industrial algorithms.

2.    Unbalanced Industrial Transformation: Mature integration in leading

enterprises and first-tier cities, while county-level regions, traditional industries and MSMEs lag far behind.

3.    Imperfect Circulation Mechanisms for Data Factors: Insufficient opening of public data, incomplete systems for data right confirmation, pricing and trading, limiting full release of data value.

6.3 Deep-seated Bottlenecks for Grassroots Implementation

1 .    Stratified Gaps in Digital Literacy: Significant disparities in AI application

capabilities between urban and rural residents, grassroots cadres and operators of MSMEs.

2.    Absent Sustainable Capital Guarantee Mechanisms: Grassroots digital

intelligence transformation relies heavily on one-time fiscal investment without long-term operation and maintenance funds.

3.    Conflicts Between Technology and Institutional Adaptation: Traditional

hierarchical administrative procedures conflict with the flat collaborative logic of digital intelligence, with unclear rights and responsibilities for cross-departmental grassroots data sharing.

4.    Poor Adaptation of Downward Algorithms: General large models fail to fit niche county and rural scenarios, with excessively high costs for customized adaptation.

6.4 universal security and Ethical Risks

Algorithmic bias, data leakage, security vulnerabilities in intelligent equipment,

structural employment shocks from AI and platform data monopolies permeate all industrial and grassroots scenarios.

7 High-Quality Development paths for Digital Intelligence Integration coordinating Global

collaboration , Industrial upgrading and Grassroots Inclusiveness

Targeting three core objectives: international competition security, high-end industrial breakthroughs and balanced grassroots inclusiveness, this section constructs a four-dimensional integrated implementation path.

7 . 1 Global Dimension: Building an Inclusive and

collaborative New pattern of Global Digital Intelligence Governance

1 .    Participate in Formulating Global AI Standards: Leverage China’s abundant

industrial scenarios to promote unified international rules for cross-border data

flows and trustworthy industry-specific large models, enhancing China’s discourse power in global digital standard-setting.

2.    Establish a Support System for Digital Intelligence in Developing Economies: Export lightweight digital infrastructure and inclusive grassroots AI solutions to narrow the North-South global digital divide.

3.    Parallel Pursuit of Self-Reliance and Opening Cooperation: Make

breakthroughs in core technologies including underlying computing power, chips

and basic models, while advancing compliant cross-border data circulation to balance security and openness.

7. 2ToP-Tier Industrial Dimension: Build an Independently controllable, Leading Global Digital Intelligence Industrial Ecosystem

1 .    Strengthen Whole-of-Nation System Research Breakthroughs: Focus on key fields such as industrial large models and general intelligent agents, improving  collaborative innovation systems linking national laboratories and leading

enterprises.

2.    Tiered Cultivation of Industrial Benchmarks: Build full-chain digital twin demonstration bases for advanced manufacturing, and national-level digital intelligence integration pilot zones for digital finance and digital trade.

3.    Improve the Market-Oriented System for Data Factors: Perfect systems for data right confirmation, trading and income distribution, open public data

resources to unleash the value of data factors.

4.    Leading Enterprises Driving Coordinated Transformation of MSMEs: Introduce fiscal subsidies and inclusive computing power policies to lower intelligent

transformation costs for micro and small enterprises.

7 .3 Grassroots Implementation Dimension: Build a Tiered, Inclusive Lightweight Downward penetration system

1 .    Tiered Construction of County-Level Shared Computing Power Infrastructure: Build unified computing centers at the county level for shared use by towns,

villages and MSMEs to reduce grassroots investment costs.

2.    Tiered and Classified Precise Implementation Schemes: Deploy full-scenario in -depth integration in developed regions, while prioritizing lightweight livelihood  and industrial rigid-demand applications in underdeveloped areas.

3.    Regular Grassroots Digital Literacy Cultivation: Establish long-term AI skills training mechanisms for village cadres, farmers and MSME operators.

4.    Improve Sustainable Grassroots Operation Mechanisms: Adopt diversified

funding models combining fiscal subsidies, enterprise service fees and village

collective industrial revenues to resolve vacant operation and maintenance after one-time infrastructure investment.

5.    Promote Age-Friendly and Agriculture-Friendly Digital Inclusive

Transformation: Simplify interactive modes of intelligent applications and support offline auxiliary services to eliminate digital exclusion.

7 . 4 Institutional Governance Dimension: Improve a secure,

controllable and Human-centered Digital Intelligence Governance system

1 .    Perfect Tiered and Classified AI Supervision Systems: Benchmark international risk classification frameworks to implement differentiated regulation for industrial, government administration and livelihood AI applications.

2.    Establish filing systems for algorithms and data security review mechanisms to guard against data leakage and algorithmic discrimination risks.

3.    Promote Two-Way Adaptation Between Technology and Institutions: Reform fragmented grassroots administrative management systems, establishing clear

rights and responsibilities for cross-departmental data sharing to break institutional barriers.

4.    Balance Technical Efficiency and Humanistic Value: Digital intelligence

integration shall ultimately target livelihood inclusiveness and common prosperity, avoiding technological alienation and overemphasis on efficiency alone.

8 conclusion and outlook

The in-depth integration of AI and digital intelligence constitutes the core thread of the new round of technological and industrial revolution, carrying strategic

global competitive height, depth in modern industrial transformation and breadth of urban-rural grassroots inclusiveness. Differentiated development paradigms

have taken shape worldwide, with technological competition, rule games and

digital divides as long-standing coexisting global patterns. Domestically, digital

intelligence integration drives high-end restructuring across manufacturing,

finance, agriculture, urban governance and digital trade, spawning new growth

points for industrial value. Deployed at grassroots units including towns,

communities and MSMEs, digital intelligence technologies can make up for

shortcomings in public services and activate grassroots industrial vitality, yet they remain constrained by infrastructure, capital, digital literacy and institutional

adaptation.

To advance high-quality development of digital intelligence integration, internal

and external landscapes must be coordinated. Externally, China shall deeply

participate in global digital governance, adhere to parallel independent innovation and openness and inclusiveness, and build a balanced and inclusive global digital development system. Domestically, a dual-track advancement system of "top-tier breakthroughs in high-end industries + tiered inclusive grassroots

implementation" shall be constructed. On one hand, breakthroughs in core

underlying technologies shall be pursued to build world-class digital intelligence  industrial clusters. On the other hand, lightweight technology penetration shall be promoted to narrow digital development gaps between urban and rural areas,

large enterprises and MSMEs. The deep integration of AI and digital intelligence will continuously foster new productive forces, providing core momentum for the construction of Digital China and Chinese modernization.

In the long run, human-machine collaboration, full-domain intelligence and

inclusive data factors will become long-term evolutionary trends of digital

intelligence integration. Future research shall continuously balance technological innovation, security governance and livelihood equity, avoiding negative effects including technological monopoly and digital differentiation, so that the

integration of AI and digital intelligence can truly serve common global development and domestic common prosperity.