OMB issues guidance on Federal Agency use of Artificial Intelligence

As of Dec. 1, 2024, federal agencies will be required to create a governance structure and meet minimum practices in order to deploy and use artificial intelligence (AI) as part of agencies daily work according to an announcement by Vice President of the United States Kamala Harris on March 28.

The announcement, which coincided with the publication of a memo by the White House Office of Management and Budget (OMB), indicates federal agencies will be required to appoint a Chief Artificial Intelligence Officer (CAIO), create an AI governance group of agency leaders, and utilize minimum measures to mitigate risk to public safety and risk to individual rights associated with agency use of AI.

The HIMSS policy team has created a fact sheet with the minimum requirements for federal agency deployment and use of artificial intelligence.

The framework, which doesn’t apply rules driven process systems whose behavior is defined only by human-defined rules to produce the same calculation when the same data is ingested, defines covered AI as:

• Any artificial system that performs tasks under varying and unpredictable circumstances without significant human oversight or that can learn from experience and improve performance when exposed to data sets.
• An artificial system developed in computer software, physical hardware, or other context that solves tasks requiring human-like perception, cognition, planning, learning, communication or physical action.
• An artificial system designed to think or act like a human, including cognitive architectures and neural networks.
• A set of techniques, including machine learning, that is designed to approximate a cognitive task.
• An artificial system designed to act rationally, including an intelligent software agent or embodied robot that achieves goals using perception, planning, reasoning, learning, communicating, decision making and acting.

To utilize covered AI tools, federal agencies must complete an AI impact assessment for each tool and real-world testing for each AI tool, to be vetted by the new CAIO and AI governance structure for each agency.

Any AI tools demonstrating unacceptable risks to safety or individual rights must be shut down and removed from agency work processes unless the AI tool meets specified exemption criteria. The OMB guidance also requires federal agencies to revalidate AI performance on an annual basis starting in December 2024.

The OMB memo also requires federal agencies to establish a CAIO within 60 days, with a deadline of May 27, 2025, and to publicly issue (on the agency’s website) the agency plan for compliance with the minimum practices for mitigating risks to safety and individual rights within 180 days (Sept. 24, 2025.)