Ep. 241 Embedding Trust into AI for the Public Sector
Feds At The Edge by FedInsider
Release Date: 03/25/2026
Ep. 241 Embedding Trust into AI for the Public Sector
Feds At The Edge by FedInsider
Trust is difficult enough in an environment with strict controls and security; AI adds dimensions that make establishing trust even more challenging in the public sector. Today, we sat down with three experts who share insights into achieving this elusive trust. Leaders must evaluate how to trust three elements: the model, the data, and the monitoring processes. Tim Willging from Rocket software suggests that model choice matters for transparency. A federal leader will need to document the system thoroughly, the training data, and known risks. One way to accomplish that is with a...
info_outline
Ep. 240 Navy Transformation: Modernization is About the Mission
Feds At The Edge by FedInsider
How do you upgrade a global fleet when failure isn't an option? This week on Feds At the Edge, we dive into the Navy's "Operation Cattle Drive," an ambitious initiative to modernize technology across hundreds of ships in the world’s most challenging environments. We sit down with experts from the U.S. Navy and Red Hat to break down the three pillars of high-stakes digital transformation: Development: Using digital twins to test systems in advance—eliminating risk where failure isn’t an...
info_outline
Ep. 239 Facing the Two-Edged Sword of AI: Securing Government Infrastructure in the Face of AI Automation
Feds At The Edge by FedInsider
AI presents such a dazzling set of opportunities that federal leaders can be tempted to dive in without careful planning. This week on Feds At the Edge, we move past the hype to explore the tactical realities of deploying AI within the unique constraints of the federal environment. Ashley Billman, Cybersecurity Analyst, Pacific Northwest National Laboratory, explains why agencies must align AI architecture with mission and security needs, whether on-prem, hybrid, or cloud environments, and highlights a critical question: what’s actually in the training data, and...
info_outline
Ep. 238 State and Local Government: Streamlining Operations to Improve Efficiency, Cost Effectiveness
Feds At The Edge by FedInsider
AI has the power to transform government operations, but only if it’s used wisely. This week on Feds At the Edge, we dive into the high-stakes world of applied AI within public agencies, where the potential for massive efficiency gains sits right next to the risk of moving too fast without a map. From Todd Sharkey of Lorain County, OH’s look at the document "drowning" in many government organizations to Brandon Ragle at Illinois Department of Innovation and Technology’s vital warnings on data privacy and...
info_outline
Ep. 237 Human + AI Prioritizing Higher Value Work
Feds At The Edge by FedInsider
I isn’t here to replace government professionals, it’s here to elevate them. Artificial intelligence is still new territory for many public sector leaders, but understanding how to apply it effectively can unlock major gains in productivity and insight. This week on Feds At the Edge, we dive into the essential transition for government professionals: offloading routine tasks to AI so humans can reclaim high-level analytical work. Dr. Nancy Washton breaks down the critical distinction between...
info_outline
Ep. 236 Overcoming Infrastructure Challenges: How Technology is Transforming Transportation.
Feds At The Edge by FedInsider
What if the solution to our most daunting infrastructure challenges isn't just more concrete, but more clarity? This week on Feds At the Edge, we explore how cutting-edge technology is turning budget constraints into opportunities for smarter, more sustainable growth. Experts from VHB, Portland Bureau of Transportation, Georgia Department of Transportation, Bentley Systems, Inc, and Reynolds, Smith & Hills, discuss how leveraging digital twins - dynamic software-based replicas of physical...
info_outline
Ep. 235 Identity Gaps are the New Insider Threat
Feds At The Edge by FedInsider
Citizens expect government services to work seamlessly on their phones, and agencies are moving quickly to adapt. But simply extending legacy systems to mobile can leave critical security and usability gaps. This week on Feds At the Edge, experts from California DMV and SailPoint break down three key lessons for building trusted, modern access. First, MFA alone is no longer enough. As attackers find ways around one-time codes, organizations are shifting from basic identity checks to stronger, continuous authentication approaches. ...
info_outline
Ep. 234 The States of the States
Feds At The Edge by FedInsider
What does a massive county-based state like North Dakota have in common with a compact, town-centered state like Vermont? More than you’d think, and the stakes for their data have never been higher. This week on Feds At the Edge, experts from North Dakota Information Technology, State of Vermont Agency Digital Services, and Palo Alto explore how states bridge geographical and administrative divides to secure their digital future. Despite stark differences in scale and structure, both faces universal challenges: managing massive data...
info_outline
Ep. 233 Cybersecurity: Building Resiliency, Reducing Redundancy
Feds At The Edge by FedInsider
Most practitioners building cybersecure environments know the basics: simplify system procedures, invest in continuous education, and elevate visibility across the network. But what happens when new mandates are applied to legacy systems? The “bolt-on” approach is often expensive, frustrating, and ineffective. This week on Feds At the Edge, experts from NETCOM, DARPA, and Axonius discuss how automation, AI, and continuous...
info_outline
Ep. 232 Harnessing AI and ML for Smarter Cybersecurity
Feds At The Edge by FedInsider
Today, we hear how to leverage the much-vaunted AI and ML technologies to make practical cybersecurity improvements for the federal government. The analysis includes comments about setting a base line, prioritizing alerts and a quick dive into the characteristics of Operational Technology (OT). BASELINE: Signature-based risk analysis has proven to be easy to deceive. Alex Maier from August Schell suggests that the solution is a move to a “behavior” based tool. In other words, see where a user's behavior varies from the norm. If that is the case, then you must know what “typical” is all...
info_outlineTrust is difficult enough in an environment with strict controls and security; AI adds dimensions that make establishing trust even more challenging in the public sector. Today, we sat down with three experts who share insights into achieving this elusive trust.
Leaders must evaluate how to trust three elements: the model, the data, and the monitoring processes.
Tim Willging from Rocket software suggests that model choice matters for transparency. A federal leader will need to document the system thoroughly, the training data, and known risks.
One way to accomplish that is with a “model card.” This document provides details on AI’s performance and training data. Model choice matters for transparency and risk documentation.
Even if we assume the data we use to train a model is good, we must consider the concept of “context of use.” One data set may be safe for one context, but not another. Users need a deep understanding of data that includes hybrid governance, legal concerns, and ethical considerations.
If we have learned anything in the past decade, it is that checkbox solutions never work. For example, if a data source is examined and deemed safe, this can change. Legacy data pipelines may not be secure, and the data may not be encrypted in transit or at rest. Continuous monitoring is mandatory for any valid AI application.
The panel also explored the role of AI in improving collaboration, data integrity, and public service, as well as the need for continuous monitoring and agile governance to ensure trustworthy AI deployments.