美国国务院-人工智能战略（英）-2023.10-14页_749kb

U.S. Department of State Enterprise AI Strategy Summary

The U.S. Department of State's Enterprise AI Strategy (EAIS) is designed to leverage artificial intelligence to enhance diplomacy, operational efficiency, and decision-making while prioritizing responsible and secure use. Adopted in October 2023, this two-year strategy (FY2024-FY2025) builds on the Department's Enterprise Data Strategy (EDS) to address the transformative potential of AI in global affairs.

Vision

The Department's AI vision is to responsibly and securely harness trustworthy artificial intelligence to advance U.S. diplomacy and shape the future of international statecraft.

Strategic Drivers

• Global proliferation of AI technologies, particularly user-friendly generative AI, is disrupting traditional diplomacy and requires adaptation.
• Opportunities for responsible AI exist in domains like language translation, foreign policy analysis, and information processing, but must be balanced with risk mitigation.
• To lead in AI adoption, the Department must integrate AI into its infrastructure, workforce, and operations while ensuring security, privacy, equity, and ethical use.

Goals and Objectives

Goal 1: Leverage Secure AI Infrastructure

• Objective 1.1: Enable AI technology integration by providing broad access to AI systems (open-source, commercial, custom) with robust security controls.
• Objective 1.2: Fully utilize infrastructure for AI adoption by enhancing computing, development, testing, deployment, and monitoring capabilities, leveraging cloud and secure data pipelines.
• Objective 1.3: Modernize AI tool acquisition through streamlined procurement processes, compliance with regulations like FedRAMP, and risk assessments.

Goal 2: Foster a Culture that Embraces AI Technology

• Objective 2.1: Provide AI training and support services to build AI fluency and facilitate responsible use across all departments.
• Objective 2.2: Develop new opportunities for AI talent by recruiting and hiring AI experts, creating new roles like data scientists.
• Objective 2.3: Promote responsible AI use through interdisciplinary training, education on risks, and adherence to core principles like data privacy and integrity.

Goal 3: Ensure AI is Applied Responsibly

• Objective 3.1: Establish and maintain AI governance and policy to oversee risk management, algorithmic bias mitigation, and compliance with legal and ethical standards.
• Objective 3.2: Broker appropriate access to AI-ready data through transparent sourcing and platforms like Data.State, ensuring data security.
• Objective 3.3: Facilitate data quality assurance by implementing robust cleaning, monitoring, and assessment processes to support reliable AI operations.

Goal 4: Innovate

• Objective 4.1: Identify opportunities for AI use cases through entrepreneurial approaches, leveraging data campaigns, partnerships, and prioritized funding.
• Objective 4.2: Facilitate responsible experimentation by creating sandboxes for low-risk testing and sharing best practices.
• Objective 4.3: Scale successful AI applications through collaboration with interagency bodies, academic, and industry partners, with clear evaluation guidelines.

Moving Forward

Implementation will involve cross-departmental coordination, led by the AI Steering Committee (AISC), Enterprise Data & AI Council (EDAC), and Chief Data and AI Officer (CDAO). The strategy complements the EDS using Data Campaigns to drive focused projects, with regular reviews to assess progress and inform future adjustments. Financial and technical support are key, emphasizing a cultural shift toward AI literacy and responsible innovation.

This strategy supports U.S. foreign policy by enabling advanced capabilities while mitigating risks, positioning the Department as a leader in trustworthy AI.