By Katie Howell
Artificial Intelligence has garnered much attention in recent months, including across the federal government. Executive orders and guidance over the past two Administrations signal bipartisan interest in developing and adopting AI technology at the federal level to ensure competitive advantage as well as mitigate certain issues with AI ethics and accountability.
The Data Foundation released “Eager for AI: Data Needs to Ensure Artificial Intelligence Readiness in the Federal Government,” a white paper developed in collaboration with the Data Coalition's Artificial Intelligence (AI) Task Force. Published in April, the paper is intended to serve as a resource for policymakers interested in AI oversight or funding needs, agencies looking to begin developing and implementing AI plans, or industry groups looking for ways to support the government.
High-quality, reliable, and accessible data feed AI technologies – it is critical that as agencies work to address the mounting pressure to realize the benefits of AI, they must first have a substantial amount of high-quality data, governed by effective principles, to deploy AI technologies and tools. Many federal agencies are taking steps to leverage AI technologies to improve internal processes and program operations, however, agencies’ strategies vary widely in maturity and alignment with their data strategies. “Eager for AI” consolidates existing strategies that federal agencies are employing that center around strong data management, governance, and technical capacity, and recommends opportunities for agencies and interagency bodies to support data needs for AI development and adoption.
The paper acknowledges key challenges that exist and identifies three key components of safe, effective adoption of AI: high-quality data, effective governance principles, and technical capacity. Numerous agencies are highlighted for their progress, with the aim to encourage collaboration and information sharing among agencies. However, agencies may lack transparency around how data are managed and used for AI implementation and the progress they are making in their own AI strategic plans. More transparency around certain data needs can similarly spark such collaboration, bringing in more perspectives to help solve challenges and help move agencies forward as they integrate AI into operations beyond pilot phases.
To facilitate collaboration and develop an approach that prioritizes data needs in federal agencies so that AI can be effective, equitable, and ethical, the Data Foundation presents various recommendations for agencies and cross-agency bodies, such as the Chief Data Officers Council. Among the recommendations, the paper suggests agencies should evaluate “AI readiness” by identifying specific steps, responsible oversight bodies, and performance metrics in AI plans; maintain an inventory of implementation progress on a single, publicly accessible website; and continue implementation of the Foundations for Evidence-Based Policymaking Act. The Chief Data Officers Council can help coordinate agencies’ AI planning and implementation efforts by publishing guidance on data standards and management practices that facilitate agencies’ ability to adopt AI technologies and developing a government-wide AI workforce training program, among other things.
Use of AI has the potential to reduce cost, ease compliance burdens, and improve effectiveness of operations and service delivery, and while agencies need not wait until their data strategies are fully implemented before considering additional aspects that are needed to prepare for AI adoption on a larger scale, data management and governance must be prioritized. By highlighting agencies that are taking encouraging steps to ensure data readiness, “Eager for AI” aims to bring data to the forefront to ensure the federal government is able to leverage the exciting potential of these emerging technologies.
Read the Report
1100 13TH STREET NORTHWEST SUITE 800WASHINGTON, DC, 20005, UNITED STATESCOALITION@DATAFOUNDATION.ORG
RETURN TO DATA FOUNDATION