This June marks 81 years since Gen. Dwight D. Eisenhower ordered the cross-channel invasion of France — a decision dependent upon hundreds, if not thousands of interrelated factors. Success depended on successful deception, suppressing German coastal artillery, preventing Panzer reserves from turning Allied forces back into the sea, and ensuring friendly forces could expand their lodgment and sustain momentum. Weather alone required careful consideration: tides, sea states, wind conditions, visibility, cloud ceilings, and illumination had to align to support airborne and amphibious operations.
Eisenhower also had to ensure that maritime forces controlled the English Channel, German supply lines and key bridges were disrupted, and landing craft, artificial harbors, fuel, and ammunition were in place. Each of these considerations broke down into dozens of subordinate indicators including the timing, readiness, logistics, and enemy movements — each critical to determining when to launch the invasion.
Fast forward to the present: if the ongoing war between Russia and Ukraine is any indication, the future of warfare will be an attritional, World War I-style conflict, layered with drones, electronic warfare, long-range precision munitions, and other advanced weapons. These newer technologies have expanded the number of variables, the volume of data, and the demand for accurate information, making decisions today significantly more complex. The scale, speed, and complexity of military operations have only increased since Eisenhower stood in Southwick House and muttered the words, “OK, we’ll go”.
Despite the need to quickly understand modern operational factors, combatant commands remain heavily reliant on hard-copy reports, PowerPoint briefings, emails, and phone calls for critical updates. These legacy processes create significant inefficiencies that are no longer sustainable or acceptable considering the fielding of advanced software and the need to increase staff efficiencies.
In the face of these challenges, U.S. European Command — where we serve — has led a holistic effort to modernize decision-making at the combatant command through two key initiatives. First, we’re leveraging commercial software to digitize our existing processes, establishing a foundation that integrates authoritative data with doctrinal decision tools enhanced by modern software. Second, European Command has deliberately embedded AI within the decision-making processes, enabling the real-time reassessment of military options, rapid evaluation of critical information, and the streamlining of predetermined actions through autonomous agents.
To be clear, empowering decision-making with new software tools is not a magic fix and requires more than simply integrating AI chatbots into daily work. This effort is about establishing the digital framework for military decisions at the highest level, requiring a significant lift on the part of the headquarters staff. To modernize decision-making, the Department of Defense should leverage the hard work occurring on the front lines of combatant commands and expand access to commercially available software to provide commanders with the speed and clarity they will need in future crises and conflicts.
Command and the Limits of Kill Chain Optimization
Since the 1990s, military leaders have envisioned a faster, more efficient force powered by emerging technology. These leaders echoed a common imperative: “To win the next war, decision-makers must see and act faster than the adversary.” Over time, the concept of decision superiority or decision advantage was reduced to its tactical basics: a technologically enabled battlefield where commanders identify and quickly destroy threats before they can react. To pursue that vision, the decision advantage problem-set was further distilled to a single use case — optimizing kinetic fires or the “kill chain.” While an admirable goal, the effort to close kill chains is not designed to tackle the complexity or scale of modern decision-making.
Command is more than connecting sensors with shooters. Command is the act of making decisions and ordering action. What the Department of Defense has prioritized instead is modernizing control — the tools and processes for directing and monitoring action once decisions are made. An assessment of some of the top innovation projects across the department, including the Joint Staff’s Joint All Domain Command and Control Program, Army’s Project Convergence, Navy’s Project Overmatch, Air Force’s Advanced Battlefield Management System, and others, reveals a consistent pattern of prioritizing improvements within a single use case while leaving other critical operational decisions largely unaddressed.
This persistent focus on the kill chain has come at the expense of the broader challenges facing the command-and-control system to make and order complex decisions, not just connecting the sensor to the shooter. While the narrow focus on technical fire control procedures may have been appropriate for past conflicts, it is insufficient for the complexity, speed, and scale of today’s conflicts.
The Complex Nature of Military Decision-Making
To improve decision-making, U.S. European Command is improving command holistically. Commanders make decisions through clearly defined decision points linked to critical events when they anticipate making a key choice concerning a specific course of action. While some decisions may focus on destroying a critical target, most revolve around allocating resources: determining how, where, and when to prioritize limited capabilities; and when to adjust task organization or boundaries, transition between operational phases, or execute branches or sequels.
Each of these decisions involves meeting specific friendly and enemy conditions before taking action. For targeting decisions, these conditions are well established and focused on specific tactical criteria such as verifying target custody, ensuring collateral damage is within acceptable limits, obtaining legal approvals, and confirming the availability of appropriate platforms and munitions — conditions constrained to the requirements and effects of a single strike.
By contrast, the decisions required of a joint force commander in large-scale conflict are significantly more complex. While striking a target may involve a handful of preconditions, committing a reserve force or launching an amphibious operation can entail hundreds of interdependent conditions, each carrying significantly greater risk and cascading operational consequences. These decisions also extend far beyond the immediate tactical effect of striking a single target, influencing force distribution, sustainment, and decisions that subordinate and adjacent commanders should make, thereby impacting the overall trajectory of the campaign.
Consider a scenario involving the massing of adversary forces along the border of a treaty ally. A joint force commander would need to balance several critical missions simultaneously: deterring the enemy, providing reinforcements, setting conditions for noncombatant evacuation operations, and preparing for a possible humanitarian disaster. Each mission would depend on a defined set of indicators, including enemy troop movements, naval deployments signaling a possible amphibious assault, and signs of airborne operation preparations.
In anticipation, U.S. commanders would need to consider how, when, and where to deploy high-readiness units like the 82nd Airborne Division, how to manage the declassification of sensitive intelligence, when to order the departure of U.S. military personnel, and where to position air and logistics assets to support a rapid evacuation. This balancing act, whether focused on the European theater, Levant crisis, or flashpoint in the Balkans or Caucasus, occurs routinely, highlighting the complexity and pace of decisions that commanders navigate.
Building the Digital Foundation
Many organizations across the Department of Defense are pursuing their own individual modernization solutions. However, meaningful transformation will ultimately occur within the staffs of joint commands, where the integration of joint and service systems directly support a battlefield commander’s decision-making.
U.S. European Command assessed that a new framework was required to modernize, as existing fires-based approaches were insufficient to meet the demands of our theater. To address this, the headquarters accelerated its transition from legacy systems to a digital workplace, built primarily on Palantir’s Maven Smart Systems. Through development and iteration with leadership, this platform has become more than a common operating picture. It’s a decision advantage environment that is rapidly transforming the command.
Just one year ago, no two combatant commands shared a collaborative and capable software platform across disparate defense networks, severely limiting the U.S. military’s ability to maintain global awareness. Today, with support from the Chief Digital and AI Office, 10 out of 11 combatant commands have Maven, enabling scaling across the global force. Others like NATO are now also adopting the software, enabling cross-combatant command and alliance interoperability at a scale not previously possible.
We began this effort with four simple rules: First, transformation occurs across all joint functions and time horizons. Second, situational awareness ought to be a standalone capability and everyone in the command should be able to understand what is happening without a briefing. Third, decision superiority requires digital decisions, enabled with AI. And finally, live data is the command standard.
In support of these rules, today European Command has brought together over 150 live data sources within its digital environment, empowering the staff to access and consume critical information through a single system. This integration is essential for modernizing decision-making, eliminating the need to pull information from multiple systems to manually correlate and validate through static products.
Training staff and senior leaders is also essential. The command teaches everyone, from new hires to experts, how to handle and understand data effectively. To accelerate this transition, European Command has hired data engineers to help supplement the staff while building talent in house. These Booz Allen engineers, working within the Maven ecosystem, are rapidly connecting information from different sources and creating AI-enabled tools to help leaders make better decisions, starting with improving how clearly the command can see the situation.
By focusing on people, processes, and technology together, European Command is creating a more efficient and clearer way for the military to make decisions. This helps European Command stay ahead even when facing complex operational problems, but it is just the first step.
Digitizing Decision-Making Tools and Workflows
Having accomplished all of this, European Command is now creating a digital twin of established military processes. By translating manual workflows into digital applications while preserving existing doctrine and procedures, the command is building a framework for integrating more advanced applications. This approach reinforces the essential orchestration layer between live data and workflows that users are digitizing, enabling the gradual introduction and integration of advanced analytical tools and AI capabilities.
Consider the decision support matrix, a doctrinal tool that breaks complex decisions into manageable parts using “if/and/then” conditions. This tool links enemy actions to friendly force requirements, creating a structured decision tree that captures key elements of the commander’s decision-making process.
10 months ago, this tool was completely manual. Staff officers tracked and processed critical information, including the status of friendly and enemy forces, unit locations and readiness, and supply statuses, by hand in PowerPoint. Today, a digital implementation of this tool has become a key component of our digitization effort, forcing multiple streams of disparate data together for action. As a result, the command is increasing the speed of decision-making by dynamically linking evolving battlefield conditions to decision points, delivering immediate insights and recommendations.
Replicating other manual reports produced for the command, Joint Staff, and others is already generating efficiency, potentially saving hundreds of hours of staff work every week. Recent experiments by the XVIII Airborne Corps have demonstrated similar efficiencies by replicating the capability of a 2,000-person fires cell with just 20 personnel, validating technology’s ability to increase effeciency without sacrificing effectiveness.
Automating Decision-Making Tools and Processes
These efforts are now allowing European Command to integrate AI within the decision-making process at deliberate points to automate complex tasks and execute predetermined actions. Current proposals for AI integrations are narrowly focused on leveraging chatbots to answer user-generated prompts and fail to leverage AI’s full potential. As Palantir’s chief technology officer recently highlighted, people “get stuck in the idea of chatbots… There’s a role for chatbots, but I think that they’re also… very limiting, actually. AI can do so much more than that.”
To ensure AI is integrated appropriately, we’re focusing on three aspects of the decision-making process to accelerate planning processes, improve critical information identification, and automate follow-on actions by employing AI-driven workflows to automatically initiate them.
First, AI can significantly accelerate planning processes by employing models to rapidly generate and evaluate multiple courses of action. The adage that “no plan survives first contact” is an understatement, as combatant commands are constantly required to create or modify plans, rendering previous ones obsolete and drawing on limited resources. In dynamic environments, decisions are necessarily revisited and adapted at speeds that exceed manual capabilities. Even when sufficient time is available to generate military options, traditional wargaming is financially burdensome, staff intensive, and still results in a very narrow and incomplete assessment. This highlights the need for an AI-enabled advanced wargaming platform, allowing military planners to continuously refine and adapt operations in response to rapidly changing conditions while significantly reducing resource demands.
Second, a modern decision-making process, powered by authoritative data and the right software tools, should enable real-time reassessment of key planning factors, as well as friendly and enemy conditions. While traditional doctrinal planning may be preferred given time and resources, it quickly becomes cumbersome in dynamic environments. Decisions developed in the controlled environment of a planning team often differ significantly from those encountered in conflict, forcing the staff to rely heavily on planning factors — imperfect facts and assumptions derived from the staff’s best estimate that are necessary to create a plan.
Instead, facts, assumptions, and other planning outputs (e.g., decision-making tools) should leverage AI agents to continuously monitor and synthesize critical friendly and adversarial information requirements. This tool could assist the staff to dynamically adapt to unforeseen or changing conditions and alert watch officers to information needed by the commander to inform decision-making. Consequently, doctrinal constructs like named areas of interest evolve beyond geographically fixed points, instead encompassing comprehensive data sets across intelligence, logistics, integrated air and missile defense, and other authoritative systems where models are trained to identify and prompt staff officers to relevant information.
Lastly, AI should also be leveraged to execute predetermined tasks, streamlining actions that would otherwise require manual input. For example, once a commander approves a course of action, AI tools can initiate follow-on tasks such as issuing fragmentary orders, updating operational overlays, tasking subordinate units to execute certain critical asks, or adjusting logistics plans. Applications like OpenAI’s Operator can execute complex multi-step operations, automatically retrieving relevant data, coordinating across systems, and executing predefined procedures. This level of automation not only accelerates response times but also reduces the cognitive burden on staff, freeing the headquarters to focus on identifying information gaps and drawing novel connections that are often overlooked.
The Path Forward
In 1944, Eisenhower‘s headquarters and logistics staff consisted of nearly 30,000 personnel, over 10 times larger than today’s typical military headquarters. While staff sizes are expected to shrink given current policy direction, the core requirement to enable a commander’s decision-making in support of national security priorities will remain.
The imperative to modernize is clear, and the tools to do so are finally available. U.S. European Command’s Decision Advantage Environment has laid the groundwork for digital modernization. Its command-and-control will only become more effective as the command expands its partnership with government and industry — like the Defense Innovation Unit’s Thunderforge project, Scale AI, and Anduril — to integrate other advanced capabilities into the command’s ecosystem. To bridge the gap between the Department of Defense’s current modernization efforts and the system needed to support the full range of decisions commanders make, the Department must build on the work already underway at the combatant commands and expand adoption of advanced software tools.
Bryan J. Quinn is a U.S. Army officer serving at U.S. European Command.
Bobby Sickler, PhD is a U.S. Army officer and director of the U.S European Command Transformation and Innovation Cell.
David Wiltse is the chief data and AI officer at U.S. European Command.
The views in this article are those of the author and not those of U.S. European Command, the U.S. Army, the Department of Defense, or any part of the U.S. government.
Image: Pfc. Myenn LaMotta
Jesus Christ is King
