In May 2023, the National Advisory Committee on Artificial Intelligence (NAIAC) released the First Annual Report of the National Advisory Committee on Artificial Intelligence, which puts forward 14 goals and 24 action recommendations for AI development around four major themes, and also plans its follow-up work.
01
Introduction
On May 4, 2022, the first meeting of the National Advisory Committee on Artificial Intelligence (NAIAC), comprised of 26 AI experts from academia, industry, nonprofits, and civil society organizations, was held. NAIAC is administered by the National Institute of Standards and Technology (NIST) within the U.S. Department of Commerce, and its mission is to advise the President of the United States and the Office of the National Artificial Intelligence Initiative on AI-related issues to guide the U.S. government in developing AI in a manner consistent with democratic values and respecting civil liberties.
As directed by Congress, NAIAC is required to submit a report to the President and Congress one year after its establishment, and every three years thereafter on the implementation of the National Artificial Intelligence Initiative Act of 2020 and related recommendations.
In May 2023, NAIAC released the First Annual Report of the National Advisory Council on Artificial Intelligence, which is NAIAC's first official report and includes four main themes:
1. Leadership of trusted AI;
2. Leadership in AI R&D;
3. Support talent development and provide opportunities;
4. International cooperation.
Around four broad themes, NAIAC proposed 14 goals covering areas ranging from logistics (e.g., strengthening White House and U.S. government leadership, coordination, and funding of AI) to innovation (e.g., creating an AI Research and Innovation Observatory). At the same time, to ensure the operationalization of this report, NAIAC has made 24 recommendations for action based on 14 objectives.
02
Main content
Topic 1: Leadership in Trusted AI
Goal 1: Implement governance of trusted AI
In January 2023, at the request of the U.S. Congress, NIST released the Artificial Intelligence Risk Management Framework (AI RMF), which provides detailed guidance on how to address the risks it poses at all stages of the AI lifecycle. NAIAC recommends that the White House encourage federal agencies to implement AI RMF or similar policies to effectively address risks at all stages of the AI lifecycle and to properly evaluate and iterate AI systems. At the same time, the adoption of AI RMF should not be limited to the public sector, and governments should encourage private sector adoption of this framework through existing mechanisms.
Action Recommendation 1: Support the adoption of AI RMF in the public and private sectors.
Goal 2: Strengthen the leadership, coordination, and funding of AI by the White House and the U.S. government
To ensure U.S. global leadership in trusted AI, the U.S. government needs to effectively coordinate and fund the work of federal agencies. In this report, we propose ways for the U.S. government to build AI leadership. Each approach can provide appropriate mechanisms to help coordinate, lead, and model responsible AI use, governance, and regulation.
Recommendation 2: Appoint and fill vacancies in AI leadership positions in the Executive Office of the President.
Action Recommendation 3: Provide funding to the U.S. Office of National Artificial Intelligence Programs (NAIIO) to fully fulfill its responsibilities.
Action Recommendation 4: Create a new Chief Responsible AI Officer.
Action Recommendation 5: Establish an Emerging Technology Council.
Action Recommendation 6: Fund NIST's AI work.
Goal 3: Establish and strengthen AI leadership in federal agencies
The U.S. government must lead by example by adopting and promoting trusted AI to facilitate agencies' strategic planning around AI, raise agency awareness of the use and regulation of AI, and increase public confidence in the federal government's commitment to trusted AI.
Action Recommendation 7: Ensure leadership and coordination of AI in each department or agency.
Recommendation for Action 8: Continue to implement the U.S. Congressional mandate and executive order on artificial intelligence.
Goal 4: Empower small and medium-sized organizations to enable the development and use of trusted AI
Trusted AI is an established goal for many public and private sectors. However, they generally lack relevant knowledge and skills. This is especially true in small and medium-sized organizations (SMOs), which have few resources or capacity to build a full department or office for the development of trusted AI. Bridging gaps in resources, knowledge, methodologies and skills requires the engagement and support of a wide range of partners.
Currently, practices, standards, and frameworks for designing, developing, and deploying trusted AI are created in organizations in a relatively ad hoc manner based on different countries, organizations, sectors, and levels of risk, and there is no clear guidance. To be sure, some organizations have developed tools, skills, and capabilities for it. For example, some non-profit organizations provide scientific expertise and data to companies committed to public welfare.
Action Recommendation 9: Establish a multi-agency working group to develop a framework for SMOs to adopt trusted AI.
Goal 5: Ensure that AI is trustworthy, legitimate, and expand opportunities
The ability to use AI to create opportunities depends largely on building and maintaining public trust. The executive order has directed agencies to "design, develop, acquire, and use AI in a manner that fosters public trust while protecting privacy, civil rights, civil liberties, and American values." ”
Action Recommendation 10: Ensure adequate resources for AI-related civil rights enforcement.
Topic 2: Leadership in AI R&D
Goal 6: Support socio-technical research on AI systems
Social technology research is critical to maintaining U.S. leadership in AI research and development. We need value-driven AI solutions, not just efficiency and cost savings. These solutions should incorporate American values such as equality, just outcomes, and equal opportunity. Such solutions should adopt human-centered design, protect human agency and dignity, and lead to positive social outcomes.
Action Recommendation 11: Develop a research base and community of experts focused on socio-technical research in the AI R&D ecosystem.
Goal 7: Create an observatory for AI research and innovation
The U.S. government plays a key role in ensuring that advances in AI bring broad benefits to society. Given the breakthroughs and accelerating diffusion of AI technologies, playing this key role is more challenging than ever. However, there is currently no agency to assess and measure progress in AI, identify gaps in AI technologies and their applications, and provide relevant recommendations to governments.
Recommendation 12: Create an AI Research and Innovation Observatory to measure overall progress in the global AI ecosystem.
Goal 8: Build large-scale national AI research resources
In January 2023, the National Artificial Intelligence Research Resources (NAIRR) Task Force approved their final report, Strengthening and Democratizing the U.S. AI Innovation Ecosystem: National AI Research Resource Implementation Plan.
The report suggests that many individuals, groups, and organizations are increasingly finding it difficult to access the U.S. AI R&D ecosystem. Large-scale national AI research resources will provide much-needed support and opportunities for under-resourced and underrepresented groups to enable trusted AI innovation.
Recommendation 13: Advance the implementation plan in the NAIRR Final Report and create large-scale national AI research resources.
Theme 3: Supporting talent development and providing opportunities
Goal 9: Modernize federal labor market data for the era of artificial intelligence
Work to modernize labor market information systems while ensuring that workforce privacy is protected and strong safeguards against data misuse. Combining AI-driven tools with real-time labor market data with appropriate investments and privacy protections not only adapts the workforce to changing work scenarios, but also liberates it.
Recommendation 14: Support the U.S. Department of Labor to modernize federal labor market data for the era of AI.
Goal 10: Expand the federal workforce with AI skills
The widespread use of AI technology in our daily lives, and the ethical and social issues arising from this make the training of AI workforce more urgent.
However, the U.S. government is severely short of digital talent, including those well-versed in AI technology, social, ethical, and policy fields. One reason is that the remuneration offered by the Government cannot compete with that of the private sector. In addition, existing civil servants' upskilling programs are not in place to make full use of internal talent. The U.S. government can't keep up with the pace of AI development and meet the demand for AI talent. While the U.S. government has successfully strengthened its talent pool by inviting technologists and other digital talent, these programs have not yet been able to provide institutions across the country with the scale of talent needed to ensure the competitiveness of U.S. AI.
Action Recommendation 15: Develop ways to prepare the current and future federal workforce for the AI era.
Recommendation 16: Train a new generation of public servants with AI skills.
Recommendation 17: Invest in the federal workforce to meet AI development needs.
Action Recommendation 18: Expand federal AI talent in the short term.
Recommendation 19: Reform immigration policies to attract and retain international science and technology talent.
Theme 4: International cooperation
Goal 11: Continue to foster international cooperation and leadership in AI
Achieving leadership in AI and gaining the support of diplomatic allies is critical to boosting U.S. economic growth, continuing to uphold democratic values and maintaining a competitive advantage in the future.
Recommendation 20: Maintain AI leadership by expanding and deepening international coalitions.
Action Recommendation 21: Internationalize NIST's AI risk management framework.
Goal 12: Create a multilateral coalition for the U.S. Department of Commerce's National Oceanic and Atmospheric Administration (NOAA) and the U.S. Department of State to accelerate efforts on AI for climate
One of the important manifestations of AI benefiting society is helping to address global challenges. While current climate models are good at assessing climate on a global scale, they are not as good at assessing local impacts. The development of artificial intelligence makes it possible to monitor the health of the planet, improve the resilience of transportation networks and supply chains, and reduce the risk of extreme weather events and climate-related disasters.
Research in related fields is making rapid progress, but it is scattered across academic institutions and private sector laboratories in various countries. U.S. allies and partners should work with the U.S. private sector and academia to advance important AI innovations faster and more effectively, such as earth-scale digital twins and sustainable computing methods, while working together to mitigate the environmental impact of AI development.
Recommendation 22: Build a US-based multilateral coalition to collaborate internationally on accelerating AI-for-climate efforts.
Goal 13: Expand international cooperation in AI diplomacy
The U.S. Department of State has identified emerging technologies, including artificial intelligence, as a key focus area and significant opportunity for U.S. diplomacy and foreign policy.
Recommendation for Action 23: Fully fund the newly expanded U.S. National Bureau of Cyberspace and Digital Policy and the newly established Office of the Envoy for Critical and Emerging Technologies.
Goal 14: Expand international cooperation in AI research and development
AI leadership is necessary to ensure that AI is built on democratic values, which requires the United States to coordinate with international allies and partners to establish rules and norms to foster a free and open environment for cooperation. One way to achieve this is through collaborative research and development with allies and partners. NAIAC recommended that the National Science Foundation (NSF), along with the U.S. Department of State, establish the U.S. Institute for Multilateral Artificial Intelligence (MAIRI) to promote AI research and development.
Action Recommendation 24: Establish MAIRI through NSF and the U.S. Department of State.
Disclaimer: This article is transferred from Meta Strategy, the original author Zhou Wei. The content of the article is the personal views of the original author, this public account compilation/reprint is only to share, convey different views, if you have any objections, welcome to contact us!
Transferred from 丨 Meta Strategy
Author丨Zhou Wei
About the Institute
Founded in November 1985, the International Institute of Technology and Economics (IITE) is a non-profit research institution affiliated to the Development Research Center of the State Council, whose main functions are to study major policy, strategic and forward-looking issues in the development of the mainland's economy, science and technology, and economy, track and analyze the development trend of world science and technology and economy, and provide decision-making consulting services for the central government and relevant ministries and commissions. "Global Technology Map" is the official WeChat account of the International Institute of Technology and Economics, dedicated to conveying cutting-edge technology information and scientific and technological innovation insights to the public.
Address: Building A, Building 20, Xiaonanzhuang, Haidian District, Beijing
Tel: 010-82635522
WeChat: iite_er