The U.S. Army is accelerating the adoption of artificial intelligence across the force through Project ARIA (Army Rapid Implementation of Artificial Intelligence), an initiative designed to rapidly field AI capabilities that improve decision-making, logistics, and mission execution.
Unlike traditional defense technology programs that can take years to deploy, Project ARIA focuses on delivering practical AI tools to soldiers within months, working directly with industry to rapidly field operational solutions.
These tools are designed to operate not only in centralized data centers, but also at the tactical edge, where soldiers must make decisions in real time, even in denied or disconnected environments.
But this shift toward distributed AI systems introduces a new and often overlooked security challenge: protecting sensitive data stored on endpoint devices.
Project ARIA is built around three major development tracks:
These initiatives reflect a broader transformation occurring across the Department of Defense. AI is increasingly being embedded into operational platforms, decision-support tools, logistics systems, and battlefield analytics.
To function effectively, these systems must often operate directly on edge devices, including:
In many cases, these devices store sensitive operational data locally so they can function without relying on cloud connectivity.
This is where data-at-rest security becomes mission-critical.
AI systems deployed under initiatives like Project ARIA rely heavily on sensitive data to operate effectively. That data may include:
If an endpoint device containing this information is lost, stolen, or captured, the consequences can be severe.
Adversaries may gain insight into:
In other words, the data powering AI systems can be as sensitive as the weapons platforms they support.
Unlike traditional IT infrastructure operating inside secure facilities, edge devices supporting military operations face significantly greater exposure.
Common risk scenarios include:
In modern warfare environments, planners must assume that devices will eventually be physically accessed by adversaries.
Security architectures must therefore be designed to protect sensitive data even if the device itself is compromised.
The NSA’s Commercial Solutions for Classified (CSfC) program provides a framework specifically designed to address this challenge.
CSfC Data-at-Rest (DAR) guidance requires two independent encryption layers protecting stored data. This dual-layer architecture ensures that a single vulnerability cannot expose classified information.
The two layers typically include:
The outer layer often combines hardware encryption with Pre-Boot Authentication (PBA). This layer:
The inner layer uses software-based full drive encryption to add a second, independent layer of protection for stored data. This strengthens the overall architecture by creating an additional barrier around sensitive and classified information on the device.
Together, these layers provide defense-in-depth protection for endpoint devices operating in high-risk environments.
AI deployments like Project ARIA dramatically increase the amount of sensitive data stored on endpoint devices.
Without strong data-at-rest protection:
CSfC DAR architectures help ensure that even if a device is captured or compromised, the data, and the AI capabilities it enables, remain protected.
This is particularly important as AI systems increasingly operate in:
In these scenarios, endpoint security becomes the last line of defense.
Programs like Project ARIA represent a major step forward in operationalizing artificial intelligence for the U.S. Army. By accelerating AI adoption and delivering capabilities directly to soldiers, the Army is strengthening its ability to operate in increasingly complex environments.
But as AI capabilities expand, so does the importance of securing the data that powers them.
The future battlefield will be shaped not just by weapons systems, but by data, algorithms, and intelligent decision-making platforms operating at the tactical edge.
Protecting that data, especially when stored on endpoint devices, is essential to maintaining operational advantage.
CSfC Data-at-Rest guidance provides a proven framework for doing exactly that: ensuring that even in the most contested environments, the information powering America’s military AI remains secure.
Protect your edge AI deployments before vulnerabilities become breaches.
Discover how Cigent secures mission-critical data at the edge.
Kelvin Quezada is a Product Marketing Manager at Cigent, where he leads strategy, positioning, and go-to-market efforts for the company’s Data-at-Rest protection solutions. With more than a decade of experience across cybersecurity deployments, product marketing, and technical enablement, he translates complex technical concepts into clear, mission-focused value for defense programs.
