The Defense Logistics Agency will initially focus on the use of artificial intelligence across three key mission areas: operations, demand planning and forecasting, and auditing and transparency.
At the same time, DLA is not waiting for everyone to be trained or for the data to be perfect.
Adaryl Roberts, chief information officer at DLA, said that by applying AI tools to their company's use cases, employees can actually clean up data faster.
“Humans aren't the ones trying to analyze data and draw conclusions, so we're leveraging AI to help curate data and make sure we're getting cleaner data and not just focusing on ChatGPT and things like that,” Roberts said on Ask the CIO. “I know it's a buzzword, but for an agency like DLA, ChatGPT doesn't solve the strategic problems we're trying to solve, so there's more emphasis on AI. For us, in those 56 use cases, a lot of it was natural language processing, a lot of it around procurement, and what I think is more standardized data, and what we're aiming for with generative AI.”
Much of this work is configuring DLA to use agent AI in the short to medium term. Roberts said DLA expects to apply agent AI to its mission areas to achieve the scale, efficiency and effectiveness benefits that the tool promises to deliver.
“At DLA, that's when we can have our digital employees work like humans and do the work at scale without having to redo the work. That's where we lose efficiency from a logistics standpoint. If we have to reorder or reship, it costs taxpayers more and slows warfighter readiness,” Roberts said at a recent DLA Industry Collider day. “From an R&D perspective, we're really looking at the tools that we have. We have native tools in the cloud. We have SAP, ServiceNow, etc. So based on the massive investment from technology, we're really focusing on the tools we have across the supply chain. What are the gaps from a technology perspective that we can't answer from a technology perspective? And we'll focus on very specific use cases that will help accelerate AI in that space. Another part of that is building AI so that it can seamlessly connect to the ecosystem.”
He added that this ultimately ensures that the technology is no longer a data stovepipe and can be integrated into a larger set of applications to be effective and mission-disruptive.
A great example of how this approach has led to success is DLA's use of robotic process automation (RPA) tools. Roberts said the agency currently has about 185 unmanned bots operating 24/7 to help DLA meet its mission goals.
“Through the Digital Citizen program, government officials are actually building bots. As a CIO, I don't want to be a roadblock. A lot of technology has advanced, and you can almost do things like entry-level coding by watching YouTube videos. Now high school kids are building bots. So I want to put technology in the hands of the experts, the people who know the business processes best. We should get that help to the warfighter,” Roberts said.
Due to the success of the bot initiative, DLA decided that adopting a commercial platform to implement AI tools was the right approach. Roberts said all of these platforms exist under the DLA Connect enterprise-wide portal.
“It's really about looking at technology, people, processes and data. How do you integrate that and track that at a high level so that you don't take on the technical debt that you incurred about 25 years ago? As a result, you need an architecture that lays out your business processes, your supply chain strategy, and how you integrate that within those business processes and overlay that with IT and processes within the IT space,” he said. “Business processes, supply chain, strategy, all of that overlaps. We're seeing that integration and interoperability moving forward. So we're creating a single portal where customers, industry partners, actual partners, internal DLAs can all communicate and see what's happening across DLAs.”
Training all employees on AI
He said this includes questions about contracts and upcoming requests for proposals, as well as order status updates and other data-driven questions.
Of course, no matter how good your tools are, if your employees aren't trained on how to use the AI capabilities or know where to get their data, the benefits will be limited.
Roberts said DLA has invested in training, from online and in-person courses to the creation of a specific “Innovation Navigator Course” that focuses on both IT and how to help companies within the agency view innovation as a concept.
“Not everyone needs the same level of training in data insights and AI analytics, depending on where they are in the organization. So we're working with HR and working with other executives in the mission area to understand what skill sets they need to support their day-to-day mission. What are their strategic goals? What is their workforce demographic? How do we train them in-person as well as online?” Roberts said. “We're not trying to reinvent the way AI and data learn, but how do we do that and incorporate things that are important to DLA going forward? We have a very strong plan for continuing education. We don't just take courses, we get trained. I think this is where we train people as soon as they join, but then they don't get any additional training for 10 to 15 years and then the technology passes by. So we keep up to date with the technology and it evolves as the technology evolves.”
Roberts said the training is aimed at everyone from DLA directors to senior mission area leaders to logistics and supply chain professionals. Our goal is to help users answer and understand how to use digital products, how to best inform AI tools, and how to deploy AI to impact their missions.
“You don't want to deploy AI for the sake of deploying AI, but you need to educate your workforce in terms of how AI can help them in their day-to-day work. And you need to educate them on how to strategically build it from a leadership perspective to achieve your goals,” he said. “Across DLA, we have trained more than 25,000 employees. All of our employees are exposed to at least an entry-level data acumen. Then they take an advanced level to really understand how to manage and lead when you have a digital-first concept. We're running a number of targeted courses for managers, where we walk through some use cases and complete them with some strategically defined priorities, so that you can better guide your people and how they're going to do it.'' Strengthen governance and operational success. ”
The courses and training helped DLA “lay the groundwork in terms of what it takes to be a digital organization, what it takes to be digital-first. Now we're in the execution and implementation stage and using these tools,” Roberts said.
Copyright © 2026 Federal News Network. Unauthorized reproduction is prohibited. This website is not directed to users within the European Economic Area.
