Seconds matter in modern combat. The Pentagon is deploying advanced data processing tools to frontline units, slashing sensor-to-shooter timelines with the intention of enabling commanders to outmaneuver adversaries instantly.
By Kris Osborn, Warrior
AI-enabled analytics, computing, data processing and transmission at the tactical edge is sweeping through the Pentagon as a high-priority Concept of Operation now transitioning from war-laboratories to the operational front. It certainly would be difficult to underestimate the tactical value of AI-enabled high-speed analysis and decision-making integrating with human commanders in seconds amid fast-moving combat circumstances. Latency, transport layer incompatibility, slower information management and sheer transmission time could impede operational efficiency, delay decision-making and ultimately determine life and death for forces facing enemy fire. With small, form-factor, AI-enabled data analysis at the point of collection, potentially life-saving decisions can be made in seconds or milliseconds, allowing forces to respond, attack or maneuver "faster" than an adversary.
Essential "speed" is the principal concept behind forward-operating AI, meaning a fighter jet, surface ship or mechanized ground formation can increasingly be fortified with the ability to "see," "verify," "analyze," "confirm" and "act" at speeds exponentially faster than what was previously possible with more traditional command and control. In clear terms, the successful integration of AI. For several years now, AI-integrated data processing has reduced a 20-min sensor-to-shooter verification timeline ... to a matter of seconds. Most recently, the ability to empower forces to that at the tactical edge is redefining maneuver formations and concepts of operation.
These command and control concepts, and the multi-domain formations necessary to enable them ... has become a massive focus for the U.S. Army, which is surging ahead with both a Next-Generation Command and Control program and a newly formed Multi-Domain Command led by the 7th ID from Army Pacific.
A key element of this concept, described by 7th ID Commander Maj. Gen. Bernard Harrington, is encapsulated in an emerging concept known as Cross Domain Contact Layer.
"How do we take multiple sensors, whether they are in the space layer, in the stratosphere, in the air, maybe on the maritime surface, or even on the terrestrial surface? How do we take that information associated with that, sense and make sense, and then be able to provide that to the joint force with an ability to be persistently forward," Harrington told Warrior in an interview.
In weapons and technology terms, this means configuring sensors on fighter jets and drones with a vastly improved ability to perform AI-enabled analytics in the air at the point of tactical engagement, or perhaps even engineering a missile or bomb itself with the technical ability to process threat specifics and adapt to changing targets or use two-way data-links to connect with a host platform. This kind of application is already taking shape through the Air Force's Golden Horde "collaborative" weapons in-flight data sharing effort.
In yet other terms, AI at the tactical edge means dismounted soldiers maneuvering to "contact" with an enemy through rugged terrain are increasingly able to operate with ruggedized, small, hand-held processors designed to bring AI-analytics to the forward edge. One such example can be seen with Curtiss Wright's emerging PacStar 810 server/processor AI-capable "box."
Engineering AI-enabled, high-speed processing power and analytics in a small, mobile, lightweight "box" form factor, is designed to bring sensing, PED (Processing, Exploitation, Dissemination), data analytics and critical information transmission from expeditionary, forward operations to command and control networks. In some instances, service members on the tactical edge will be empowered with the data they need to "strike," "defend" or "maneuver" in response to enemy activity themselves in a matter of seconds. In yet other instances, the PacStar 810 might "transmit" organized, time-sensitive and highly relevant information to command control centers or integrated "hubs" to optimize the decision-making process. As a small, 20-pound transportable forward-operating "server" application with AI-capable computer processing
Cross Domain Contact Layer
With these kinds of tactical scenarios in mind, Harrington described CDCL as an essential element of a layered, multid-domain fight in which relevant, time-sensitive data needs to transmit at speeds "faster" then an adversary can respond.
Harrington described it as "taking that information and then being able to pass it through layered, agile formations, whether they be offensive, or whether they be defense, whether they be kinetic, or whether they be non-kinetic. How are we able to take any sensor and be able to tie it to the best formation to achieve an effect on behalf of the joint force. Then, merging those two together, we can use adaptive and Agentic AI to get to decision quality information across the vast amount of data that is out there, and allow commanders to make decisions faster than any adversary," Harrington said.
The idea, as Harrington explained, is not simply to move data but move "decision-quality" data across formations. Within this context, Harrington emphasized that the Cross-Domain Contact Layer and Next-Generation Command and Control leverage AI for data management, analysis and pure "speed."
"With our ability to leverage AI, it's not about technology, it is about the ability for commanders to make decisions and achieve effects on behalf of the Joint Force alongside our allies and partners at the greatest speed possible. We leverage AI and Agentic AI quite a bit for two things; One is to really identify changes. AI to look at what changes are occurring in the environment and be able to prioritize information based off commanders' priorities. For example, perhaps we're in an environment now where there are more sensors and more data than really any human staff can consume.
Something like PacStar 810 could, given Harrington's description of Concepts of Operation in the Pacific, prove valuable in dispersed formations spread across a large land-air-sea territorial expanse. Forward nodes, engineered with an ability to perform analytics at the point of collection on the tactical edge, could potentially bring otherwise disparate or disaggregated pools of data together in a relevant tactical way.
"We need to operate at the scale and within the scope required in this theater; it becomes even more important as we are distributed across the entirety of the Pacific region, across the entirety of the first island chain and understand that breadth of the environment, AI and Agentic AI are critical to doing this,” Harrington said.
Tactical Implications
Curtiss-Wright weapons developers appear to have engineered the PacStar 810 with a mind to the Army CDCL concept of operation. A transportable edge computing device could connect data from otherwise disconnected sensor layers, a task described as critical to CDCL by Harrington. A Curtiss Wright essay on the PacStar 810 cites some of its technical specs, including high-speed network connectivity and one TB DDR5 memory. Essentially, PacStar 810 could use high-end signal processing to counter an attacking drone swarm from the "very tip of the spear," enabling near "real-time decision making" related to countermeasures and "defensive" sensor-to-shooter coordination. Threat specifics could potentially be verified by networked sensors, sent to the PacStar 810 and then verified against an AI-enabled data base to confirm threat specifics and also recommend or "pair" sensors with the effectors and countermeasures necessary to destroy a fast approaching threat.
