AI Update: Trump Administration Releases American Artificial Intelligence Initiative’s One Year Annual Report – Lexology

Original article was published on artificial intelligence

In February 2020, the Trump Administration released the American Artificial Intelligence Initiative’s One Year Annual Report, detailing the Administration’s progress since launching its “American Artificial Intelligence Initiative” by signing Executive Order 13859 on February 11, 2019. The Administration’s Report highlights areas where it identifies progress in advancing the United States’ competitive position in fostering advancement in AI technology, including limiting new regulations, increasing funding and tracking of federal AI research and development, increasing access to federal data and computing resources, and promoting international collaboration.

Limitation of regulations and development of guidance

The Report reiterates the Administration’s policy that federal agencies should consider ten high-level AI principles when proposing new regulatory or non-regulatory approaches to private sector use of AI technology. Specifically, the report points to the White House’s Office of Science and Technology Policy (OSTP) draft Guidance for Regulation of Artificial Intelligence Applications, released on January 7, 2019, which lays out these ten principles. The Guidance document also highlights some non-regulatory alternatives that may be effective for AI, such as policy guidance and frameworks, pilot programs and experiments and voluntary standards. The Report highlights agency actions over the past year consistent with OSTP’s approach, for example the Department of Transportation’s (DOT) Ensuring American Leadership in Automated Vehicle Technologies principles, which set out ten voluntary principles for the large-scale deployment of automated vehicles, as well as the Food and Drug Administration’s proposed regulatory framework for AI-based software as a medical device and its draft guidance for clinical decision support software.

Increased funding and tracking of federal AI R&D investments

The Administration has directed heads of federal agencies to increase their budget allocations for AI research and development and annually report investments through the Networking and Information Technology Research and Development Program. The Report notes that the FY 2020 NITRD Budget Supplement is the first agency-by-agency reporting of unclassified, non-defense federal investments in AI R&D, which totaled $973.5 million in FY 2020. The report also notes that President Trump’s FY 2021 budget requests large increases in non-defense AI funding, including increases in funding for the National Science Foundation’s (NSF) National AI Research Institutes, the Department of Energy’s Office of Science supercomputing programs, and the National Institutes of Health (NIH) for research into chronic diseases. The Report highlights the Administration’s Select Committee on AI as an ongoing locus of inter-agency coordination, as well as a continuing source of advice to the Administration on the United States’ AI priorities, including partnerships with industry and academia.

Increasing access to federal data and computing resources

Following Executive Order 13859’s call for federal agencies to identify new opportunities to increase access to federal data and models, the White House Office of Management and Budget (OMB) created the Federal Data Strategy to serve as a framework for Federal agencies seeking to manage and share data. The Strategy directs agencies to improve their documentation of their data and model inventory and make available hardware and computing power that will assist the private sector in the further development of AI. For example, the Report indicates that the NSF and the DOT’s Federal Highway Administration are collaborating on researching privacy techniques that would allow datasets to be accessed remotely, and the NSF and NIH are both experimenting with solutions to allow enhanced access to their computing resources and data.

Promote international collaboration

The Administration continues to emphasize the role of international collaboration to protect the United States’ competitive edge in AI. The Report identifies the OECD Principles on AI, as well as the G20 AI Principles, as the type of guidance on the development and use of AI the Administration sees as a worthy model of international collaboration. The Report also highlights the Administration’s prioritization of bilateral agreements and its goal of seeking agreements in line with the 2017 US-UK Agreement and 2018 US-France Joint Statement on science and technology cooperation.

The One Year Annual Report demonstrates the Trump Administration’s continued focus on AI technology development as an important theme of its economic and foreign policy. The Administration plans on avoiding regulation in this area, relying instead on high-level voluntary frameworks issued by federal agencies. The Administration also aims to increase funding for AI research and development, and to make the fruits of this public research available for utilization by civil society, highlighting recent efforts by DOT, DOE, NSF, and NIH. To the extent possible, the Administration hopes to advance these governance principles internationally through multilateral and bilateral agreements.