At this year’s Potomac Officers Club Army Summit, the Seekr-sponsored panel centered on the transformative role of agentic AI in military logistics and supply chain optimization. The all-star panel was led by Guy Beougher, Senior Advisor to Seekr and Vice President of Logistics at Cypress International, the panel also featured Adarryl Roberts, the CIO of Defense Logistics Agency (DLA); Jenn Swanson, the VP of Digital & Next-Gen Solutions at Cypress International; Robert Thurston, the former Army Director G-46; Emil Vogeler, the Executive Director of Logistics Policy and Programs for Defense Logistics Agency (DLA); and Carrie Wibben Kaup, the President of Exiger.

As defense operations grow more complex and contested logistics becomes the norm rather than the exception, the Department of Defense (DoD) is actively exploring how to leverage AI to drive speed, accuracy, and effectiveness across its fragmented logistics landscape. Here are the top five takeaways from the panel discussion.

1. Agentic AI moves from automation to autonomy

Unlike traditional automation, agentic AI introduces autonomy and decision-making capabilities to workflows. It’s not just about processing faster, it’s about acting smarter. agentic AI systems can reason, coordinate, and execute processes from end-to-end without constant human intervention. For the Defense Logistics Agency (DLA), this shift means evolving from siloed legacy systems to integrated, intelligent supply chains.

Imagine an AI agent automatically predicting a backorder and initiating procurement, redistributing global stock, or alerting relevant teams—all before the issue becomes visible to the end-user. This is the future of logistics: AI-driven systems that don’t just tell us there’s a problem—they solve it in real time.

2. Data harmonization is mission-critical

Panelists universally agreed on one thing: AI is only as good as the quality and state of the data it works with. Today, the DoD’s logistics data is siloed across multiple legacy platforms, many configured differently, even within the same agency. A good goal is creating a data fabric and normalizing master data. Agentic AI can help accelerate data harmonization, ingesting structured and unstructured data, curating data, eliminating, or detecting inconsistencies, and building a foundation for more reliable predictive models. Without AI, supply chain forecasting will remain stuck in reactive mode, especially in complex, and contested environments.

3. Human-machine teaming is critical

Agentic AI isn’t about removing humans from the loop and automating everything, but it is about giving humans better teammates. Speakers described it as a “digital co-pilot” that surfaces the right information, coordinates behind-the-scenes workflows, and proactively suggests actions. For example, customer service agents at DLA currently must swivel between twenty-four different systems to answer a single query. With agentic AI operating in the background, humans could have immediate access to comprehensive answers, across systems and data siloes.

4. Simulation and scenario planning at machine speed

Contested logistics, especially in Indo-Pacific theaters, require agile decision-making based on real-time information that is always in flux. Agentic AI is a breakthrough here too, not only for faster execution but also for rapid, adaptive scenario modeling.

Traditional “what-if” supply chain simulations are cumbersome and require pristine data and manual work. Agentic AI, by contrast, can work with messy, incomplete datasets to identify some supplier risks, transportation vulnerabilities, or parts shortages in real-time. More importantly, these systems are goal-seeking and adaptable, allowing for scenario planning at the speed of conflict, not weeks or months later, but as the situation unfolds.

5. Security, trust, and keeping pace with adversaries

The most pressing opportunity and challenge is using agentic AI to enhance supply chain risk management. It’s no longer enough to know where your parts come from; it’s about understanding who is behind those suppliers. Obfuscation by adversaries is increasing, making it harder to detect investments in suppliers, or influence from hostile nations, in seemingly benign supply chains.

As one speaker noted, agentic AI must be part of the continuous vetting process—not only surfacing potential risks but flagging them before they impact operations. And as the panel emphasized, the DoD must design systems that are “as commercial as possible, military as necessary” to stay adaptable as threats evolve.

Final thoughts

Agentic AI is not a distant vision; it’s arriving now, with early deployments happening across defense agencies. However, as the speakers candidly admitted, technology alone won’t win this fight. Success will require better data, better integration, and most critically, better alignment between commercial technology partners and defense procurement practices.