How Agentic AI in Government Can Transform Federal Operations

How Agentic AI is Used in the Workforce

Agentic AI technology links humans and agents in complex business tasks with multiple steps, according to Jason Andersen, vice president and principal analyst at Moor Insights & Strategy.

“Agents are a class of applications that take AI models, workflows, and data sets from a wide range of places, and they can interact with other applications,” Andersen says. You can delegate tasks to an AI agent, and it will only report back after it accomplished the goal or it gets stuck.

With AI agents, you outline a series of steps and specify the order of actions, Andersen says.

You also note an objective, such as delivering a report to a manager or solving a customer problem. 

You also specify the timing of actions, like having an agent run reports Monday through Friday or Tuesday through Thursday, he says.

Federal agencies can scale up agents when they are run in a cloud environment to add more capabilities during peak times, Andersen says.

“AI agents take the paperwork and friction out of government interactions, enhance decision-making by surfacing real-time insights, speed up permitting and licensing, automate administrative work like benefits eligibility checks and case updates, and free up human workers to focus on the highest-priority constituent needs,” Tatum says.

ACCOMPLISH THE MISSION with federal IT infrastructure modernization.

How Agencies Are Piloting Agentic AI in Government

Customer service applications are a key use case for AI agents at agencies including the Department of Labor and the United States Postal Service, as AI agents do away with the hold time that customers experience when calling customer service.

“AI agents can actually take on the entire process of solving a customer problem from beginning to end,” Andersen says.

In addition, AI agents allow the federal government to modernize old applications, which have been a hindrance to government agencies, according to Andersen.

“Sometimes, that technology is very expensive to upgrade and migrate,” Andersen says. “This has really been a game changer when it comes to saving money on these less glamorous projects.” 

Air Force Deploys Missionforce Agentic AI

The Air Force is using Salesforce’s Missionforce National Security to transform operations and modernize mission readiness, Tatum says.

“This work is also helping them establish a scalable foundation for future AI deployments,” he adds.

Missionforce National Security will connect the data, systems and workflows for airmen and guardians in a single operational view, Bill Pessin, senior vice president of national security at Salesforce, previously told FedTech.

In addition, the Air Force is piloting Salesforce Agentforce Public Sector, which is a digital workforce of intelligence AI agents for public sector missions. The Air Force is “exploring how AI agents can automate complex, time-consuming workflows and sharpen decision-making in the field,” Tatum says.

Further, Andersen sees the military using vision models, in which an AI agent can execute an action based on what it sees, he says.

The Labor Department Turns to Agentforce

In March, the Department of Labor announced it would use Agentforce to modernize its National Contact Center and expand the support it offers the American workforce and retirees. A Department of Labor agent will collect intake details, open formal cases and escalate cases autonomously to workers when citizens request it.

The VA Uses Missionforce and Slack AI

The Veterans Health Administration deployed an agentic operation system incorporating Slack AI to more than 150 VA medical and outpatient centers. The platform will save thousands of staff hours, allowing workers on the front lines to spend less time on paperwork and free up more time for patient care, according to Tatum. 

In addition, the Department of Veterans Affairs awarded Salesforce $1.6 billion as part of a three-year Agentic Enterprise License Agreement to deploy AI agents across VA operations using Missionforce. Tatum says the agents will modernize care and service delivery and provide around-the-clock support to veterans.

“It will also free VA staff of administrative burdens so they can spend less time navigating systems and more time serving veterans,” Tatum says.

Guardrails and Rules Governing Federal Use of Agentic AI

In February, the National Institute of Standards and Technology launched an AI Agents Standards Initiative within its Center for AI Standards and Innovation to make agentic AI systems interoperable and more secure. CAISI aims to create an ecosystem of AI standards and protocols by working with other federal partners such as the National Science Foundation. This collaboration between NIST and interagency partners will also foster open-source protocol development and maintenance for AI agents.

Meanwhile, in June, the White House issued an executive order that called out the use of AI agents by bad actors as part of security breaches. That means that federal agencies will need guardrails in place to monitor how agents are used.

“These models have now gotten to a level of sophistication where it’s a very realistic threat,” Andersen says. “The model doesn’t govern the agent; it operates independently. So, you need a set of good policies to make sure that the agent itself is behaving in a way you want it to.”

A federal agency may put guardrails in place to ensure that a cost estimate for a job doesn’t extend past a reasonable amount or the amounts of other vendors, Andersen suggests.

How To Evaluate Agentic AI Readiness for Federal Deployment

Evaluating agentic AI readiness involves examining the context as well as how all of the different data sources feed into the agent, according to Andersen.

Tatum adds that agentic AI readiness means having a unified, secure data foundation so the agents can reason across an entire enterprise, such as a federal agency.

“The agencies getting the most out of AI made the architectural decision first: Consolidate onto a unified platform, establish a single authoritative view of your data, then layer AI on top,” Tatum says. “That’s what allows agents to drive real mission outcomes rather than just automating tasks in isolation.”

To evaluate AI readiness in a federal agency, Tatum asks a department if the data is clean and unified, and if the agency has a security and compliance infrastructure to support deployment at scale.

“If the answer to those questions is yes, you’re ready to build; if not, that’s where we start,” Tatum says.

Click the banner below for the latest federal IT and cybersecurity insights.

Building an Agentic AI Roadmap for Your Agency

As a first step of a federal agentic AI roadmap, understand the requirements of what you are trying to build, Andersen advises.

“You don’t necessarily have a hardwired framework or process or workflow,” Andersen says.

“There might be variation there, so that means you need to do a lot more testing and a lot more piloting.”

Tatum advises that a road map should include a conversation with workers on what aspects of a workflow feel repetitive or low-value. He also recommends beginning with “high-value uses cases” that have clean data with easy-to-measure outcomes.

“Gather feedback, prove the value, then expand,” Tatum suggests. “What starts as a focused deployment becomes a blueprint for the entire organization.”

He also stresses that human oversight should be part of an agentic AI roadmap as a “feature” rather than a “constraint” to produce agentic AI that’s “not just efficient, it’s accountable and built to last.”