Government AI Responsibility: An Overview
In a bid to enhance service delivery and improve the citizen experience, the US Government is accelerating the adoption of Artificial Intelligence (AI) as a core component of its service delivery strategies. This shift is driven by the Government Service Delivery Improvement Act, the Integrated Digital Experience Act, OMB directives M-25-21 and M-25-22, and Executive Order 14179.
The focus is on striking a balance between innovation and responsibility, with the key steps for implementing responsible AI in government revolving around three core pillars: accelerating innovation, building AI infrastructure, and promoting responsible AI adoption and coordination.
To accelerate innovation, regulatory barriers and red tape are being removed, interagency coordination bodies are being formalized, and AI procurement toolboxes are being established. A "try-first culture" is being fostered through regulatory sandboxes, and federal employees are being equipped with AI literacy and training.
Critical sectors, particularly the Department of Defense, are prioritized for AI adoption, with a focus on streamlining operational workflows and partnerships with cloud providers for priority computing access in emergencies. Procurement guidelines are being updated to ensure AI systems used by the government are objective and free from ideological bias.
Open-source AI models, scientific datasets, and an AI evaluations ecosystem are being encouraged to support transparency, innovation, and trustworthiness in federally used AI technologies. Agencies are also being urged to prioritize data protection, ensuring personally identifiable information is processed in ways that safeguard privacy.
Agencies are also being advised to maintain the ability to intervene when needed, proactively preventing problems with continuous testing, human review, and oversight. AI should be trained to ensure the right outcomes and intervene when necessary to prevent unintended results.
The IRS, for instance, rolled out a smart callback option during the 2023 tax filing season to help up to 95% of callers avoid long hold times. Similarly, a state DMV is considering using AI to detect website issues and proactively trigger outbound contact center outreach via smart scheduling tools.
Federal agencies are also using AI to uncover and address the root causes of calls, as demonstrated by the Social Security Administration's addition of a "return to my saved application" button. The VA, since 2017, has been using AI to better understand the veteran experience and identify veterans at risk of self-harm, suicide, and homelessness.
Consumers prefer shorter wait times and positive satisfaction ratings during customer service interactions, according to a 2024 Medallia study. Most consumers find waiting up to five minutes on hold reasonable but prefer call-back options if available, as indicated by another Medallia study published in 2023.
The adoption of AI in government is aimed at improving everyday processes for citizens and employees, making service delivery more efficient, and raising the bar on the citizen experience. However, it's crucial to remember that this transition requires a commitment to responsible AI, promoting transparency, accountability, and security.
[1] White House Office of Science and Technology Policy, "Executive Order on Maintaining American Leadership in Artificial Intelligence," 2020. [2] White House Office of Science and Technology Policy, "National AI Research Resource Task Force Report," 2021. [3] White House Office of Management and Budget, "Memorandum M-25-21: Establishing a Federal Strategy to Responsibly Develop and Deploy Artificial Intelligence," 2021. [4] White House Office of Management and Budget, "Memorandum M-25-22: Promoting Efficient and Effective Federal Spending to Advance the Federal Government's Use of Artificial Intelligence," 2021.
The reimagined workforce, including the federal workforce, is being equipped with AI literacy and training to foster innovation and streamline operational workflows. The federal government is promoting the responsible adoption of AI technology, with a focus on enhancing service delivery and protecting citizen privacy through initiatives such as the removal of regulatory barriers, prioritizing data protection, and using AI to proactively identify and address issues.
