国际商会国际仲裁院-实现包容性人工智能（英）-2025_19页_250kb

Achieving Inclusive AI Summary

Executive Summary:

• AI offers transformative potential for innovation and productivity but risks uneven benefits unless developed inclusively.
• Key building blocks for inclusive AI include reliable infrastructure, equitable access to data and compute power, digital skills, multilingual models, ethical frameworks, and supportive policies.
• Collaboration between governments and industry is crucial to expand AI access globally, particularly in developing countries and underserved communities.

Key Findings:

1. Building Blocks for Inclusive AI:

   • Infrastructure & Connectivity: Reliable energy, high-quality broadband, and accessible digital infrastructure.
   • Access to Data & Compute: Availability of diverse, high-quality, and responsibly governed datasets, along with computational resources.
   • Skills: Digital literacy, generalized AI skills, specialized technical skills, workforce transition support.
   • Local Innovation: Encouraging homegrown AI solutions, linguistic inclusion (multilingual models), and supporting the start-up ecosystem (including SMEs).
   • Ethical AI & Governance: Robust ethical frameworks, privacy protection, cybersecurity measures, and harmonized international standards.
   • Policy Environment: Enabling policies (national strategies, public sector use, regulations), international cooperation, and initiatives focused on Global South access.

2. Role of Governments:

   • Infrastructure Investment: Promoting energy security, building digital infrastructure (broadband, data centers), and updating grid management.
   • Access to Data & Compute: Establishing data commons, data-sharing frameworks, data stewardship, improving public data.
   • Skills & Education: Implementing national AI education strategies, promoting AI literacy, vocational training, workforce transitions.
   • Public Awareness: Running educational campaigns to foster understanding and trust in AI.
   • Policy & Regulation: Developing national AI strategies aligned with international principles, enhancing procurement policies for AI, ensuring privacy and data protection laws, formulating cybersecurity standards, managing data flows, addressing intellectual property rights in AI context, harmonizing standards.
   • Public Sector Use: Leading by example with AI adoption.
   • Public-Private Partnerships (PPPs): Fostering innovation hubs, supporting start-ups.
   • International Cooperation: Participating in capacity-building initiatives and partnerships (like Current AI, AI4D).

3. Role of Business:

   • Innovation: Driving inclusive AI development and deployment through ethical and responsible practices.
   • Market Solutions: Creating practical, market-driven AI products and services.
   • Private Initiatives & Collaboration: Forming multi-company collaborations and working within multistakeholder platforms.
   • Supporting SMEs: Lowering barriers and promoting fair competition.

4. Importance of Inclusivity for Global South & Underserved Communities: Ensuring these regions can benefit from AI is vital for equitable and sustainable development. This requires targeted investment in infrastructure, data, skills, and capabilities.

5. Cross-Cutting Themes:

   • Ethics: Upholding human rights, fairness, transparency, and accountability.
   • Skills: Continuous learning for workforce adaptation.
   • Regulation: Clear, non-prohibitive policies fostering innovation while protecting citizens.

Conclusion/Recommendations:

International cooperation is crucial to bridge the AI gap by focusing together on inclusive energy and infrastructure deployment; providing flexible and adaptable access to relevant data, skills, and locally relevant AI applications; enabling flexible and adaptable collaboration, public-private partnerships; with appropriate human rights, privacy, data protection, and security safeguards.

Policy imperatives include:

• Foundational Infrastructure: Reliable electricity, clean energy, high-quality broadband connectivity, and sustainable data centers.
• Accessible Education & Data: Interoperable public data and national strategies aligned with international ethical frameworks.
• Inclusive Innovation: Supporting local AI development, linguistic inclusion, and a start-up environment.
• Regulatory Systems: Promoting early regulatory systems, especially around data governance, privacy, and cybersecurity, integrated with harmonized AI standards.