← The Daily Scoop Podcast
The Daily Scoop Podcast
The CIA looks to commercial technologies to fuel innovation
00:23:34
storically tumultuous year for federal employees …
Loading summary
Transcript50 lines
- [00:01]
A
Today on the Daily Scoop Podcast from the Scoop News Group brought to you by hpe. The CIA looks to commercial technologies to fuel innovation. It's Tuesday, November 18, 2025. Welcome to the Daily Scoop Podcast where you'll hear the latest news and trends facing government leaders. I'm the host of the Daily Scoop Podcast, Billy Mitchell. Thanks so much for joining me. Now let's dive into the day's top headlines. A historically tumultuous year for federal employees didn't dim the public's pre shutdown view of government services, according to a new survey that largely credited tech adoption for the positive perceptions. The 2025American Customer Satisfaction Index federal government study released on Tuesday found citizen satisfaction with federal government services at a 19 year high with a score of 70.4 on a.0100 scale scale, which is a 1% jump from 2024. The survey of 6,914 randomly chosen respondents was conducted before the longest government shutdown in US history, but ACSI's director of research emeritus emphasized that the results still reflect real momentum in improving how citizens experience federal services. Forrest Morgeson, an associate professor of marketing at Michigan State, in addition to his role at acsi, said that the introduction of AI is making a huge impact and such advantage advancements signal a future where government services can be more responsive and accessible to all. Many of the highest ranking federal agencies in customer satisfaction were lauded for their implementation of technologies, including usda, the State Department and the Small Business Administration. Across agencies, the survey found little change in how the public broadly views government processes, websites, customer service and the availability of information. But it did reveal a notable improvement in call center satisfaction, a 5% boost that the ACSI linked to callback systems, AI powered virtual agents and voice bots. It remains to be seen if the shutdown in the Trump administration's decimation of the federal workforce halts the post pandemic surge in positive customer service reviews for the US Government. Now, moving on to other news, the National Institutes of Health didn't ensure that the entity housing personal information of over 1 million people, including bio samples, implemented proper cybersecurity protocols, according to an internal watchdog. In a report publicly released on Friday, the Department of Health and Human Services Office of inspector general made five recommendations for the security of the all of Us program, a database of diverse health information from 1 million participants that's meant to aid research after finding weaknesses, according to the report. While the award recipient operating the program's data and research center implemented some cybersecurity measures, NIH failed to ensure other controls were addressed. The report found that NIH didn't ensure that the awardee, which wasn't identified appropriately, limited access to the program's data and didn't communicate national security concerns related to maintaining genomic data or data related to DNA. It also failed to ensure that weaknesses in security and privacy were fixed within a timeline outlined in federal requirements. The audit was initially conducted by the Inspector General due to the threats that cyber attacks and the potential exposure of sensitive information can pose to the agency's programs. The watchdog's objective was to scrutinize the access, security and privacy controls of the program. The all of Us program began in 2016 after it was authorized by Congress under the 21st Century Cures act and was designed to provide researchers with a database that could inform thousands of studies on a variety of health conditions, per its website. As it stands, the participant data included in the database includes 470,000 electronic health records and 607,000 biosamples. For more news at the intersection of the federal government and technology, make sure to visit fedscoop.com there's an ongoing massive shift across the federal government towards buying and adopting commercial technologies. While the intelligence community's sensitive classified mission set might not seem like a good fit for such commercial tech, that actually isn't the case, according to Ryan Klotz, deputy CIO for IT Enterprise at the CIA. Earlier in the fall I interviewed Klotz at our annual Fed Talks event about the intelligence community's push to commercial the barriers to adoption and how those agencies in the IC are working with industry to ensure success. Now let's go to that interview with Ryan Klotz.
- [04:47]
B
Welcome to Fed Talks.
- [04:48]
C
Yeah, thank you for having me.
- [04:48]
B
Yeah, it's a fantastic event so far and I'm really excited to dig in with you about commercial tech in the ic. Obviously this year we've heard quite, quite a bit about commercial tech and really the focus in the Trump administration through a couple key executive orders in that direction. So starting off with a really just basic question to set the foundation, how is the IC approaching this? Obviously there's some nuance in differentiation in the IC and the intelligence community, then the rest of the government, how they have to deal with some of the adoption of technology. So how does that look as you think about commercial tech?
- [05:23]
C
Yeah, we're excited about the executive order that was issued in April that really defines the administration's policies around a commercial first strategy. When we think about our primary mission at CIA, obviously foreign intelligence collection, but really all Source analysis, the ability to make sense of data, the volume and velocity of data that we are asked to understand and analyze, to provide information to our policymakers requires us to have at our fingertips cutting edge commercial technologies. The ability to custom build and code applications in that sort of digital domain is impractical anymore. And so we view it as mission essential to be able to have at our fingertips and within reach of our mission collectors, our mission analysts, the ability to leverage these commercial technologies. It is just critical to what we do every day.
- [06:20]
B
And just as a quick follow up, the benefits that come with that speed, commonality, what's the key ones there?
- [06:26]
C
Yeah, when you think of, as we've heard numerous times this morning, AI is really that kind of game changer application. You could use other sort of big data analytics and other capabilities, but the ability to leverage these frontier models, we are not going to have the ability to build our own model. So how do I integrate those rapidly within our environments? How do I democratize access to AI so that it's not us as technologists that have to deliver those capabilities that we enable those sort of general users to connect to their data, to summarize, to create insight and analytics from that data. Right. That, that rapid integration of that technology is essential to that success.
- [07:10]
B
That makes sense. CIA has a very specific mission. Let you guys focus on that mission. Let the technology or the technologists focus on the technology.
- [07:17]
A
I'm curious, you know, I'm sure there.
- [07:19]
B
Are barriers in this adoption. I'd be curious what they are in terms of how you focus on introducing commercial tech into your domain that, you know, again might be different from elsewhere in government.
- [07:29]
C
Our industry partners that are in the audience probably can appreciate that our acquisition processes are not as transparent as they need to be. The lift of technology from the unclassified domain to the classified domain is not as easy as it should. I think we've taken that on very aggressively in the last several months, really focusing on trying to provide clarity for those new technology startups that have traditionally found those barriers to entry into the IC just far too hard to overcome. We've really adopted an idea and a mindset of prototype to production, leveraging unique acquisition authorities to partner with new technology vendors to prototype solutions in the unclassified domain and then based on success of those prototypes, create pathways and avenues to bring those into the production environment as quickly as possible.
- [08:21]
B
Ryan, you mentioned sort of your industry partners and acquisition process, but that's a two way street and I'm sure industry can play a part in making this More successful.
- [08:29]
A
So what is the role of industry in this and how are you working.
- [08:33]
B
With those vendor partners to make sure that you'll be successful beyond just making a deal, buying something that they're there for the long run to ensure that the rollout is actually a successful adoption?
- [08:44]
C
Sure, I'm going to steal a line from a previous mentor. We need our vendor partners to be great and engaged beyond the point of sale. When we often buy technology, we struggle to fully optimize and implement the full capabilities of those technologies. Whether it's because the technology company doesn't have the cleared resources on hand, whether it's our traditional systems integrators don't have a foundational understanding of the technology. We're just often in a position where we're just not optimizing and really reaping the benefits of these investments in new technology. So we need to get licensing to professional services. We need great partnerships with our traditional SI partners. We saw that with great success in our initial cloud endeavors, where we saw great investment in training and development of skills in those areas to help us realize sort of the opportunity of cloud transformation. We need to continue to see that over a broader spectrum of technology. So be great, be engaged beyond the point of sale, help us be successful. Because if it's not, if we're not able to really implement to that like 90th percentile of fully optimized implementation, you know, one we run the risk of ripping and replacing, we just, we're not getting the benefit of those new technologies within our mission domain. So maybe that's not the solution. We gotta get another one, which then sets us back or, you know, the government says, hey, I have more gaps than the tool can cover. I have to buy another tool. And right now we drive inefficiencies and in this sort of constrained financial environment that we're likely headed to, we're just not gonna have the capacity to continue to buy more tools to fill those gaps. And often those are perceived gaps because they haven't fully implemented the technology right. And so I need great partners. I need help really realizing the investments that we're making. The benefit of the investments we make. I need to be optimized to the greatest extent possible. And for our sis that are partners that we rely on day to day for that staff and technical augmentation, we need help getting to kind of that full optimized realization of the technology and capabilities, capabilities we're seeking to integrate into our pipelines.
- [10:50]
B
Well, I'm sure there's a number of People in the room that are excited to hear that and you know, are ready for that sort of participation.
- [10:55]
A
You know, as we close out, I.
- [10:58]
B
Think it's maybe a little bit of a misnomer or maybe it's just me that thinks of it this way because the ICCIA has a really specific and sensitive mission. You would think about the tech stack being customized or bespoke, but that's not really the case. You guys have sort of been on the, the front lines, the sort of cutting edge of commercial tech with the adoption of commercial cloud with C2S and C2EMO more recently. So as you think about that foundation you've set over the last decade plus in the cloud, how do you continue to look to commercial tech to drive mission value with that strong foundation? And what does the future look like? You mentioned AI, but I'm sure there's a number of other ways you're looking to build off of that.
- [11:36]
C
Yeah, sure. I had a great opportunity in 2014 to join our cloud program at launch. And I got to see the evolution at CIA and the intelligence community, the adoption of cloud. But if we pull back even before 2014, I think one of the key realizations we had was there's not a lot of differentiated heavy lifting in the way CIA would rack and stack servers or implement storage. The analytics, the data that we would collect is certainly sensitive. But why is CIA using mission resources to do kind of the fundamental rack and stack? And so we saw C2S and now C2E as that foundational infrastructure. You know, the strategy of cloud first or cloud smart, you know, is essential to what we're trying to do. But over time that's really evolved beyond cloud and infrastructure as a service is we now possess for the use in the intelligence community. And now in partnership with broader sort of dod, we have access to that foundational commercial cloud infrastructure. What we have in our private instantiation of a public cloud on our classified networks is the same cloud that lives in the commercial domain. And what we now have access to is those commercial products and technologies that are natively developed within cloud environments. We have that platform. It should make the lift from the unclassified domain to the classified domain much easier. And that's really the exciting opportunity we have in front of us. How do we take my cloud, the digital transformation that accompanied that, to further move up the stack? How do I get to no code and low code solutions? How do I leverage our partnerships with our cloud service providers to bring in frontier and open source AI models? How do I democratize access today to AI to support our mission outcomes. Cloud is that foundation. We're continuing to reap benefits from that transformation and investment to move up the stack and rapidly deploy kind of best in class commercial capabilities to our mission.
- [13:31]
B
Sounds like a very exciting endeavor, maybe a challenging one, but I think a bright future ahead for the CIA. Let's give Ryan a round of applause.
- [13:38]
D
Fantastic.
- [13:39]
B
Thanks so much for sharing your insights.
- [13:40]
C
Yeah, thank you very much.
- [13:42]
A
For more on federal commercial technology adoption, make sure to visit fedscoop.com Also in this episode, HPE Networking Chief AI Officer Bob Friday joins SNG Host Wyatt Cash in a sponsored podcast discussion on how as agencies can leverage cloud and AI to build more automated, secure and mission ready networks. This segment was sponsored by hpe.
- [14:10]
D
Public sector organizations continue exploring ways to modernize and automate their networks and network security. I'm Wyatt Keshwaskoop News Group and joining us to share his perspective on the evolution of network modernization in the public sector, particularly in light of Hewlett Packard Enterprise's recent acquisition of Juniper Networks, is Bob Friday, Chief AI Officer for HPE Networking. Bob, thank you so much for joining us. Welcome to the program.
- [14:39]
E
Yep, thank you for having me. Great to be here.
- [14:42]
D
So Bob, what are the most pressing challenges that you're hearing from your federal IT customers right now regarding the need to modernize and automate networks and at the same time when many are facing a new era of funding and staffing constraints?
- [14:59]
E
Yeah, you know why? I think you know what I hear from some of our largest federal customers? The same thing I hear from my commercial customers. Right. You know, they need methods to basically automate a much more growing complex networking environment. They know that there's value of moving to the cloud. They know that it's cloud is a much better way to develop and maintain software. But the challenges they're having is around security. The government has some very special security requirements. I'm sure you're familiar with them fips with cloud, we have now Fedramp, you know, to get to secure cloud. So I think we're hearing the same thing from federal customers that we hear from our commercial customers. With the additional requirement we have to meet new security requirements to move to the cloud.
- [15:43]
D
And I'd be interested to hear what technology trends or developments are shaping or even driving your organization's approach to infrastructure and network modernization that federal executives should be aware of.
- [15:57]
E
Yeah, you know, I think, you know, when I started Mist 10 years ago, I would say AI was still kind of a concept. I think right now I think there's a couple things and you move into a, what I call a cloud AIOps paradigm. One is, you know, you definitely have to move into more of a real time day two operation mode, you know. So when I started mist, I knew I would have to go build a whole new real time cloud architecture. But the other thing I found is organizationally I really had to tie my data science team to my customer support team. And so I think any CIO or anyone running a, you know, an IT organization, they're in the middle of this how to basically embrace all the agentic AI technology coming down the pipe. You know, there's all the build buy partnering decisions have to be made. But I think for most CIOs, they have to realize there's an organizational change. You know, as we get into this new agentic AI programming, you know, how we develop and automate software is changing and you really have to get your domain experts involved to know if you're making progress, which really requires kind of an organizational thought process, you know, whether they use a vendor or whether they do it internally.
- [17:06]
D
Absolutely. I'm also curious, can you update us on what HPE's recent acquisition of Juniper will mean to federal customers and what they should expect in the coming year now?
- [17:18]
E
Yeah, so I mean I think if you look at the mission, I think the mission has not changed. You know, joining HP and Juniper coming together, it's still the same vision and mission of trying to get to a self driving network. You know, how do we leverage the cloud and the syntic AI technology to bring together a more automated network? I think what you're going to see with going forward right now, there's two vectors that are really going to accelerate this drive to self driving. One is the agentic AI technology itself that's going to accelerate us into automating things that we haven't been able to in the past. And two, we have two market leading data science teams coming together. You know, I have an Aruba data science team, a Juniper Ms. Data Science team and now we're able to get everyone to push up the same rock up the hill. So those are the two things that are going to accelerate this whole self driving networking vision.
- [18:09]
D
And from the customer's point of view, can you talk about maybe some use cases where federal agencies are benefiting from Hewlett Packard Enterprises expertise and how that's likely to be enhanced with the acquisition of Juniper?
- [18:23]
E
Yeah, I think the other trend we're seeing is if you look what we're doing with Zoom and teams, right? You're starting to see a convergence of networking with the app monitoring space, right? So federal customers, commercial customers, you know, typically we're moving to more of a paradigm of client to cloud. You know, we still have to keep all the networking infrastructure up and running. AP switches, routers, all that stuff has to be green. But what's more important is that critical business experience, right? Whether it's a critical patient experience in a hospital or retail experience, that is what's really changing right now inside of the federal space is how do we actually go from monitoring a client to cloud experience? Which really means like, you know, zoom teams. We have customers right now. Zoom and teams are kind of like canary in the coal mine. And the question is, what happens when your customers start complaining about their zoom teams? And they're usually the first to have problems. And so what we have now is federal customers and even commercial customers where they have zoom teams problems that need to be sorted out. You know, with this new models we can build these large experience models. We can now narrow that down to which layer of the onion is causing the problem. You know, is it a client, is it a wireless, is it the LAN or is it the wan? So I think that is an example where we're starting to see both client, you know, app monitoring, network monitoring converging together to solve a real problem.
- [19:50]
D
I'd like to dive a little deeper into the security aspects of network modernization. In light of the time and cost involved in upgrading agency networks, the rapidly evolving threat landscape that we often talk about, what should agencies focus on to enhance the security capabilities of their wired and wireless networking infrastructure?
- [20:13]
E
Well, I think, you know, federal governments, federal customers have, you know, government customers have kind of two aspects. You know, they have the same challenges that commercial customers doing going to zero trust networking are going to SASE models. But immediately what most federal governments customers are dealing with is really this post quantum cryptography issue that's coming down the pipe, right? Because you know, I think BY Is it 2035, you know, federal government here in the US is going to require almost all networks to be post quantum cryptography safe. There's another regulation 2030 just to be transport, you know, 1.3 transport safe. So I think that is what's immediately. I talked to a lot of my federal customers right now, like one of the largest healthcare federal customers right now. And you know, their question is how much is it going to cost to go to be post quantum safe? You know, what is it going to take to get AP switches and routers up to a level that, you know, they don't have to worry about, post, you know, quantum attacks.
- [21:13]
D
Absolutely. And then finally, what parting recommendation would you offer public sector leaders about how to overcome inherent challenges with AI native network solutions and maybe take greater advantage of them as well?
- [21:30]
E
Yeah, I mean, I think the first challenge, as I said, you know, when you say the word AI, on the other side of that coin is cloud. You know, so when we move into these AI native networking challenges, inherently you want to get the data somewhere where you have a lot of compute and storage, and that ultimately means you want to get it to a public cloud or somewhere we can actually do something with that data, you know. So my challenge is usually around FedRamp. Get to know Fedramp, get to know the different levels of Fedramp, and start that journey sooner than later.
- [22:05]
D
Well, terrific. Bob Frady, thank you so much for joining us here today and sharing your perspectives on just the evolution of networking, wired and wireless, and the importance of gearing up for this rapidly evolving AI age that we're all trying to figure out. So thank you so much for joining us.
- [22:23]
E
Yep, thank you for having me and, you know, topic dear to my heart. So feel free to reach out to LinkedIn. You want to continue the discussion.
- [22:31]
D
Terrific.
- [22:31]
E
Thanks again.
- [22:57]
A
Thanks so much for tuning in to another episode of the Daily Scoop Podcast, available on all podcast platforms. If you've already rated the podcast on.
- [23:04]
B
Your platform of choice, thanks so much.
- [23:06]
A
Hi Ran.
- [23:06]
B
Ratings and good reviews of the show.
- [23:08]
A
Help more people to find it.
- [23:09]
B
The Daily Scoop Podcast is a production.
- [23:11]
A
Of the Scoop News Group in Washington, DC. Adam Butler and Carlin Fisher help put the show together and the entire Scoop News Group team contributes. We'll be back tomorrow with more top headlines.
- [23:22]
B
Until then, I'm your host as always, Billy Mitchell. Thanks so much for listening.