美国
Front cover: The National Artificial Intelligence R&D Strategic Plan: 2019
Dear Colleague letter from Michael Kratsios, Deputy Assistant to the President for Technology Policy
Table of Contents
Executive Summary
Introduction to the 2019 National AI R&D Strategic Plan
Sidebar: 2019 Update: RFI responses inform the 2019 National AI R&D Strategic Plan
AI as an Administration Priority
Development of the 2019 Update to the National AI R&D Strategic Plan
An Overview of the 2019 Update to the 2016 National AI R&D Strategic Plan
Structure of this 2019 Update to the 2016 National AI R&D Strategic Plan
AI R&D Strategy
Strategy 1: Make Long-Term Investments in AI Research
2019 Update: Sustaining long-term investments in fundamental AI research
Advancing data-focused methodologies for knowledge discovery
Enhancing the perceptual capabilities of AI systems
Understanding theoretical capabilities and limitations of AI
Pursuing research on general-purpose artificial intelligence
Developing scalable AI systems
Fostering research on human-like AI
Developing more capable and reliable robots
Advancing hardware for improved AI
Creating AI for improved hardware
Strategy 2: Develop Effective Methods for Human-AI Collaboration
2019 Update: Developing AI systems that complement and augment human capabilities, with increasing focus on the future of work
Seeking new algorithms for human-aware AI
Developing AI techniques for human augmentation
Developing techniques for visualization and human-AI interfaces
Developing more effective language processing systems
Strategy 3: Understand and Address the Ethical, Legal, and Societal Implications of AI
2019 Update: Addressing ethical, legal, and societal considerations in AI
Improving fairness, transparency, and accountability by design
Building ethical AI
Designing architectures for ethical AI
Strategy 4: Ensure the Safety and Security of AI Systems
2019 Update: Creating robust and trustworthy AI systems
Improving explainability and transparency
Building trust
Enhancing verification and validation
Securing against attacks
Achieving long-term AI safety and value-alignment
Strategy 5: Develop Shared Public Datasets and Environments for AI Training and Testing
2019 Update: Increasing access to datasets and associated challenges
Developing and making accessible a wide variety of datasets to meet the needs of a diverse spectrum of AI interests and applications
Making training and testing resources responsive to commercial and public interests
Developing open-source software libraries and toolkits
Strategy 6: Measure and Evaluate AI Technologies through Standards and Benchmarks
2019 Update: Supporting development of AI technical standards and related tools
Developing a broad spectrum of AI standards
Establishing AI technology benchmarks
Increasing the availability of AI testbeds
Engaging the AI community in standards and benchmarks
Strategy 7: Better Understand the National AI R&D Workforce Needs
2019 Update: Advancing the AI R&D workforce, including those working on AI systems and those working alongside them, to sustain U.S. leadership
Strategy 8: Expand Public–Private Partnerships to Accelerate Advances in AI
Abbreviations
Responsible parties and contributors
About the main bodies responsible for the strategy contents, and copyright information
Back cover