
Hosted by FedInsider · EN

When it comes to air-gapped networks, there is an assumption that this is the ultimate protection for Federal networks. This week on Feds At the Edge, we speak with three federal experts who share their observations that challenge these assumptions and bring potential cracks in this seemingly impenetrable security measure to light. We'll explore how air-gapped systems can still be exposed through compromised supply chains, why patching and upgrades become more difficult without internet connectivity, and how manual data transfers using USB drives and external media can introduce new vulnerabilities. The conversation also examines critical questions around Controlled Unclassified Information (CUI), software allowlisting, and strategies for blocking unauthorized executables. Tune in on your favorite podcast platform to learn why an air gap alone is not enough, and what agencies must do to protect their most sensitive systems.

Innovation often follows a familiar cycle: rapid expansion, followed by consolidation. Federal agencies are experiencing this firsthand as AI tools and models proliferate across their environments. This week on Feds At the Edge, technology leaders explore the challenge of rapid innovation and how agencies can regain control before the chaos of disorganized, duplicate data and tool sprawl compromises their missions. Jim Smid, Federal Solutions Architect, Palo Alto Networks, explains how observability now extends beyond servers and IoT devices to include AI models, tools, and their interactions with agency data. Jessica Souder, AIRS Specialist / Director, Public Sector, Palo Alto Networks, emphasizes that agencies should carefully evaluate new AI tools, validating them in controlled environments, and establishing governance before deployment. Tune in on your favorite podcast platform to hear why our panel of experts agree that any AI deployment should focus on the agency mission, valid data, and that all projects should commence with visibility.

AI is putting ransomware on steroids, and on this week's episode of Feds At the Edge we examine several approaches to reducing the impact of malicious actors through advanced protection strategies and smarter cybersecurity budgeting. Michael Dent, Retired CISO with Fairfax County, shares how he takes cybersecurity training to the next level with what he calls "Challenge Point," rewarding employees for identifying signs of potential attacks. Glendon Schmitz, Virginia State Corporation Commission, discusses the importance of showing leadership the direct financial impact of an attack when seeking successful budget approval. Akamai Technologies' Douglas Holland explores the emotional tactics malicious actors use to pressure unsuspecting users into complying with urgent requests. Tune in on your favorite podcast platform for more on this and ransomware-as-a-service, the long-term effects on public trust, leadership accountability, and the growing need for initiative-taking governance and budgeting.

In a recent Forrester report on federal technology, more than half of the respondents revealed they face significant negative impact from budget constraints, while a whopping 86% expect a breach. This week on Feds At the Edge we sit down with experts from Centers for Medicare & Medicaid, and Carahsoft to discuss practical strategies for securing mission-critical systems, maximizing existing technology investments, and prioritizing the areas of greatest risk. We also bring in a special guest from Forrester who shares how agencies can maximize limited resources by using shared enterprise services, streamlining overlapping AI security tools, and prioritizing critical areas such as network security, data protection, and incident response.

One common theme among technology leaders discussing AI deployment is the need for a "definitive source." Without trusted data, none of AI's conclusions are truly actionable. This week on Feds At the Edge, two technology leaders share practical strategies for building trustworthy AI environments while strengthening security and compliance. Jared Pane from Elastic discusses why organizations can't blindly trust AI outputs, highlighting both familiar data quality issues and emerging threats like hidden "white-space text" embedded in datasets that can carry malicious code into AI repositories. Meanwhile, Chris Bunton from Texas Dept of Agriculture, explains how automation can simplify security and compliance, including the implementation of a policy bot that provides employees with 24/7 guidance on compliance best practices. Tune in on your favorite podcasting platform for practical advice for agencies beginning their AI journey: start small, solve clear problems, and build a collaborative culture that supports long-term success.

Preparation is everything when it comes to AI in government. This week on Feds At the Edge, experts from the Library of Congress, Fortinet Federal, Pryon, Dell Technologies and Data Evolution share practical insights on moving beyond "lift and shift" approaches, selecting the right tools, and ensuring data is truly AI-ready. From treating AI tools as "perishable as tomatoes" to rethinking legacy data and modern migration strategies, the conversation highlights the importance of strong data foundations, thoughtful implementation, and continuous learning. Tune in on your favorite podcasting platform for insights on how successful AI adoption depends less on the tools themselves and more on preparation, collaboration, and informed decision-making.

Data is the lifeblood of AI, and when it's incomplete, biased, or inaccurate, your results will be abysmal. This week on Feds At the Edge, we sit down with Geoff Schaefer of Leidos and Susan Laine from Quest to discuss how to overcome the common pitfalls and data challenges of federal AI adoption. Our guests explore the shift toward data platforms that include curated, reusable data products that improve visibility, assign trust scores, and clarify data sources and permissions. Tune in on your favorite podcasting platform for insights on governance, interoperability, and ethical use, as this conversation highlights how strong data foundations are essential to making AI reliable, scalable, and mission-ready.

Professional sports reward those who perform under pressure. In federal technology, the stakes are even higher. This week on Feds At the Edge, we sit down with experts navigating the realities of securing federal systems in an environment where threats are constant and decisions carry real consequences. Gary Barlet, Public Sector CTO for Illumio, poses a critical question: "Do you have a plan for what happens when a malicious actor breaches your system?" He explains how segmentation can stop attackers from being able to "land and expand," and why "real-time" observability is essential to understanding actual network dependencies. Mark Mitchell, Federal Security Architect at Netskope, highlights a familiar challenge: "If this alert is valid, what is my next step?" He introduces "Active User Coaching," which helps guide users with warnings and safer alternatives when risky actions occur. Together, the conversation also explores aligning zero trust strategies with federal guidance and the importance of assuming breaches from the start. Tune in on your favorite podcasting platform to hear how effective federal cybersecurity requires planning, real visibility, and clear action in the moments that matter most.

In order to determine its value, Federal Technology leaders know it's all about the workflow when it comes to the possibility of incorporating AI. Once they do, having a deeper understanding of the distinction between traditional and generative AI, as well as the importance of data cleanliness and security will be key for an effective adoption. This week at Feds At the Edge, we outline the steps agencies should know and take to take as they prepare to expand into the newest digital landscape. From defining terms, to ensuring clean and secure data, to identifying and removing potential bias, we'll explore what the new lifecycle of information should look like. Tune in on your favorite podcasting platform to hear how streamlining the data process can help optimize the value that AI can bring to the table.

We have all seen the power of AI; the question is, can it be implemented in a secure federal environment? This week on Feds At the Edge, we sit down with three experts who are proving that powerful AI can be deployed securely in federal environments, without compromising compliance or mission outcomes. Leaders from the Library of Congress, Idaho National Lab, and Keysight share how responsible implementation, curated data, and private AI models are unlocking transformative results. You'll hear how agencies are navigating sensitive data repositories, building secure AI ecosystems, and fostering collaboration to accelerate innovation, while keeping privacy and security front and center. Tune in on your favorite podcast platform for a practical look at how federal leaders are securely scaling AI from experimentation to real-world impact.