Government Agencies Are Eager To Deploy AI, But Are They Ready?

Originally published by Government Technology Insider, Nina Jacobs.

Government agencies are eager to deploy AI throughout their organizations, but whether those agencies are actually prepared for AI-driven changes varies significantly. AI’s explosion into the marketplace means that many government organizations have been forced into a reactive stance to keep up with rapid technological advances. Now that agencies are identifying the applications and use cases that will support their missions and goals, they’ll need a more strategic approach to building out AI infrastructure. To help government agencies become AI-ready, Campbell Palmer, Vice President of Solution Architecture, Public Sector at Lumen, explains three essential elements for agencies to build the backbone of AI success.

AI and Data Governance

Government agencies, especially at the federal and state levels, hold copious amounts of sensitive data. Governance strategies need to include frameworks for securely incorporating their data into AI to protect both its accuracy and privacy. “The last thing you want to have in a closed loop AI infrastructure is bad data,” said Palmer. “Inaccurate data produces inaccurate results, impeding operations and decision-making.” Governance strategies also need to include frameworks for data validation to ensure that data is not being mistreated or misused by AI models. Palmer continued, “That’s not a problem with the data; it’s AI making the wrong decisions and even generating misinformation or hallucinations.” Procedures for auditing information produced by AI models are essential to ensure the data remains relevant.

Secure and Flexible Connectivity

When talking about AI infrastructure, data centers and repositories are often the first things that come to mind. But a modern network infrastructure is critical for connecting data, AI and users in two primary ways. The first is connectivity into secure infrastructure, such as from data center to data center or from data center to data repository. “AI models and data often reside in multiple facilities, requiring extremely high-capacity network integration,” Palmer said. “Whether you go with optical wave, 100 Gbps, 400 Gbps or even dark fiber, that’s the underpinning of the interconnect points for the various systems to communicate.” Dark fiber—fiber that is already installed but currently unused—presents an opportunity for agencies to use pre-installed cable to create, control and manage high-speed network capacity. The increased security, reliability and control provide agencies with a network that meets the mission without excessive network latency and cost. A secondary means of connectivity is necessary for applications and users to access AI information. This pathway should have security controls through a private network or secure means of IP networking to ensure only known, trusted users can access the AI engines and the data they produce.

AI Requires Agility and a Network that’s On Demand

With government agencies still in the early stages exploring AI use cases and the technology itself still in its infancy, it’s essential that the infrastructure agencies build today is agile and scalable. “There’s a high likelihood that even as you’re scaling network capacity to meet immediate needs, it may not be sufficient to meet medium or long-term needs,” Palmer said. “So, agencies need to be agile in how they approach building their network infrastructure. The AI-ready network is able to scale to meet the future needs without requiring expensive rip and replace or other massive financial investment to add the capacity that’s going to be necessary.” Network on-demand functionality, such as the capabilities ordered through a Network-as-a-Service platform, lets organizations take advantage of burstable service to keep costs in line while still allowing capacity for large compute loads when necessary. Networking services based on usage are flexible and scalable enough to support agencies through myriad objectives and needs.

Conclusion

We know secure and consistent access to data is essential for government agencies to successfully deploy AI use cases that enhance mission delivery and meet citizen expectations. While robust governance, data hygiene and security are foundational, investing in a scalable and agile network is equally critical—one that can support today’s demands and adapt to future mission-driven AI applications that have yet to be imagined. Interested in hearing from government IT leaders as they prepare their organizations, team and constituencies for the AI era? We’re sharing their stories each week on The CIO Connection.

Explore insights on our Lumen Public Sector Solutions page—including our latest white paper, Building AI-Ready Networks—for a closer look at the infrastructure, strategy and innovation behind our Lumen^® Private Connectivity Fabric^℠ and how it’s integral to infrastructure modernization.

Learn more

 This version of the article has been reprinted with permission from Government Technology Insider and author Nina Jacobs.

This content is provided for informational purposes only and may require additional research and substantiation by the end user. In addition, the information is provided “as is” without any warranty or condition of any kind, either express or implied. Use of this information is at the end user’s own risk. Lumen does not warrant that the information will meet the end user’s requirements or that the implementation or usage of this information will result in the desired outcome of the end user. All third-party company and product or service names referenced in this article are for identification purposes only and do not imply endorsement or affiliation with Lumen. This document represents Lumen products and offerings as of the date of issue. Services not available everywhere. Lumen may change or cancel products and services or substitute similar products and services at its sole discretion without notice. ©2025 Lumen Technologies. All Rights Reserved.