Educating Command for Strategic Uncertainty: Rebalancing Professional Military Education in the Age of AI, Drones and Multi-Domain War

Scope: This assessment examines the United States professional military education system, its statutory and institutional foundations, the pressure created by artificial intelligence, autonomous systems, drones, data-intensive command architectures and contemporary high-intensity warfare, and the strategic question of whether technological adaptation should displace, coexist with, or be integrated into the study of strategy, history, political purpose and command judgment over the 2026–2031 horizon. Testo incollato

---

Executive Summary / BLUF

The central issue facing American professional military education is not whether officers should master artificial intelligence, autonomous systems, contemporary wargaming and multi-domain operations; those competencies are becoming indispensable, while the current 2026 National Defense Strategy and Department-level policy explicitly place renewed emphasis on warfighting effectiveness, operational relevance and the preparation of leaders for contemporary conflict. Media Defense

The more consequential institutional question is whether the additional technical requirement is financed by reducing the intellectual preparation required to connect military action to political purpose, because existing U.S. law defines joint professional military education around national military strategy, planning at all levels of war, doctrine, command and control, force development and operational support rather than around platform-specific proficiency alone. Codice degli Stati Uniti

The strongest public evidence supports integration rather than substitution: AI, simulation, data science and technologically sophisticated wargaming can be embedded inside strategic education while preserving history, theory, political analysis and human judgment as the framework through which technological capability is interpreted and employed. The Army War College is already publicly documenting experiments that use AI within strategic-history instruction and AI-assisted wargaming rather than treating technology and strategic education as mutually exclusive categories. War Room – U.S. Army War College

This distinction matters because technical superiority does not itself determine strategic success; senior military education exists precisely at the institutional boundary where operational capability must be translated into theater strategy, national objectives, joint campaigning and politically sustainable outcomes, responsibilities explicitly reflected in statutory JPME Phase II requirements. Codice degli Stati Uniti

The principal risk over 2026–2031 is therefore not technological modernization itself, but curricular crowding: if AI, autonomous warfare, classified simulation and contemporary operational problems are added principally by removing strategic history, political-military analysis and theoretical education, the United States could strengthen near-term technical fluency while weakening the intellectual adaptability required when assumptions about the next war fail.

The decisive policy question is consequently how the Department measures educational effectiveness: if success is defined narrowly as familiarity with current systems and operating concepts, curriculum will tend toward tactical immediacy; if it is defined as the capacity to exercise judgment under political, technological and strategic uncertainty, the appropriate model becomes an integrated strategic curriculum in which technology is a means of inquiry rather than the organizing purpose of senior officer education.

EU Cyber Resilience Act Reporting Rules Go Live as ENISA’s Single Reporting Platform Changes the Operating Model for Product Cybersecurity

The Pentagon’s Education Test: How to Absorb AI Without Training Generals for Yesterday’s Technology

The United States is redesigning professional military education at the same moment that artificial intelligence, autonomous systems and digitally connected warfare are shortening the life cycle of military knowledge. The contradiction is immediate: the 2026 National Defense Strategy demands a force better prepared for contemporary conflict, while 10 U.S.C. Chapter 107 still assigns senior military education a broader purpose — national strategy, joint planning, campaigning and command — that cannot be reduced to technological proficiency. The governing choice for the 2026–2031 period is therefore not whether the war colleges should adopt AI, but whether they can do so without allowing rapidly depreciating technical knowledge to consume the scarce educational space used to produce strategic judgment. The institutions that solve that problem will strengthen command; those that confuse modernization with curricular substitution will produce officers highly fluent in the present and less adaptable to the war that does not resemble it.

Washington has already decided that military education must become more operationally relevant

On 6 February 2026, the Department ordered a reassessment of professional military education and external graduate programs under the memorandum “Rebuilding the Warrior Ethos in Professional Military Education,” tying educational investment more explicitly to the production of capable warfighters and future senior leaders. On 27 February 2026, a further memorandum on Senior Service College opportunities described these institutions as mechanisms for producing critical thinkers able to integrate multi-domain joint operations and serve at strategic level. The institutional direction is clear: Washington wants more measurable military relevance from expensive senior education.

That does not amount to an order to turn war colleges into advanced technical schools. CJCSI 1800.01G, dated 15 April 2024 and still governing officer professional military education in 2026, defines an assessment architecture built around institutional effectiveness and demonstrated learning outcomes. The implication is more demanding than simple modernization: a new AI course, simulation laboratory or classified exercise has value only if it produces a better commander, planner or strategic adviser. Course count is an input; strategic performance is the output.

The distinction is visible in the Air War College’s 34-semester-hour Master of Strategic Studies curriculum for Academic Year 2026. Foundations of Strategy receives 6 semester hours, Great Power Studies 5, Regional Security Studies 5, Global Campaigning 4, Air, Space and Cyber Power in the Future 4, research or electives 4, Strategic Leadership 2, Contemporary Strategy 2, and introductory and capstone requirements another 2. Technology is present, but it is embedded inside a wider architecture of strategy, geography, campaigning and command. That distribution offers a practical answer to the reform debate: modernization does not require replacing strategic education if emerging technology is integrated into the decisions senior officers are already required to make.

The numbers show why the curriculum cannot remain technologically static

The pressure for technological adaptation is not rhetorical. In the FY2026 defense budget request, the Department identified $13.4 billion for autonomy and autonomous systems, including approximately $9.4 billion for unmanned and remotely operated aerial vehicles, $1.7 billion for autonomous surface systems, about $210 million for autonomous ground vehicles, and approximately $1.2 billion for enabling autonomy capabilities and software. These are not experimental-edge figures; they demonstrate that autonomy has entered force-structure and resource planning at scale.

The defensive side is moving just as rapidly. The FY2026 request for counter-unmanned systems reached $3.187 billion, against $2.247 billion enacted for FY2025, an increase of $940 million, or approximately 41.8 per cent in one budget cycle. The Department’s own Strategy for Countering Unmanned Systems describes commercial innovation, AI, autonomy and networking as forces allowing both states and non-state actors to generate increasingly consequential military capabilities. Senior leaders who do not understand the economics, software dependencies, electronic-warfare vulnerabilities and replacement cycles of these systems will be unable to make credible force-design or campaign decisions.

The same logic applies to command. The Department’s Data, Analytics, and Artificial Intelligence Adoption Strategy identifies superior battlespace awareness, adaptive force planning, faster and more resilient kill chains, resilient sustainment and more efficient enterprise operations as principal outcomes. Combined Joint All-Domain Command and Control, or CJADC2, has already moved beyond concept language: in February 2024 the Department announced delivery of an initial minimum viable capability intended to improve information and decision advantage across the force. The relevant educational issue is therefore no longer whether officers will command inside AI-mediated information environments, but whether they understand enough about those systems to challenge them.

AI changes the commander’s job before it changes the commander’s authority

The most consequential shift is cognitive rather than mechanical. An AI-enabled headquarters does not simply perform the same staff process faster; it redistributes work among people, sensors, databases, networks and algorithms. The U.S. Army War College’s August 2026 study “Fighting with Data” describes AI-enabled mission command in precisely those terms, arguing that artificial intelligence redistributes cognitive labor across humans, data, tools and operational processes. That alters the point at which senior judgment enters the chain.

A conventional staff asks whether available reporting supports a recommendation. An AI-enabled staff must additionally ask what data trained the system, which information it excluded, how it ranked competing evidence, whether the model is operating inside validated conditions and how quickly an error could propagate once automated recommendations enter operational systems. More information can therefore produce less clarity if machine-generated outputs multiply faster than commanders can validate them.

This is why DoD Directive 3000.09 on Autonomy in Weapon Systems remains central to the education problem. The directive requires autonomous and semi-autonomous systems to be designed so commanders and operators can exercise appropriate levels of human judgment over the use of force, while also requiring appropriate testing, reliability, effectiveness and suitability. Human judgment is not eliminated by autonomy; it moves upstream, into mission definition, constraint setting, model supervision, intervention criteria and the decision to accept or reject machine output.

The war college is becoming a laboratory as well as a school

The institutional transformation is already visible. The U.S. Naval War College conducts more than 50 gaming events each year and operates a dedicated wargaming facility of approximately 111,000 square feet, including a 180-seat auditorium and multiple configurable gaming spaces. In November 2025, it introduced 100 international military students from 58 countries to contemporary wargaming methods, demonstrating that gaming has become not only an American educational tool but also a mechanism for multinational professional development.

In June 2026, two Naval War College student officers spent more than 300 hours developing classified AI-enabled software intended to improve wargaming and combat assessment through more repeatable analysis of military courses of action. That project represents a deeper change than adding a technology elective: students were building analytical infrastructure inside the educational institution itself.

The Army War College is following the same trajectory. During Academic Year 2026, selected seminars experimented with large language models to accelerate free-form wargame adjudication, allowing students to spend less time on rigid mechanics and more time on campaign decisions. On 1 October 2026, the College also documented an elective using AI in the study of strategic thought while requiring students to challenge outputs and preserve human judgment. The emerging model is not “AI instead of strategy”; it is AI inserted into the process through which strategy is taught, tested and criticized.

The biggest risk is not AI failure but educational dependency

A senior officer who cannot use AI will become increasingly ineffective; an officer who cannot reason without it may become dangerous. That distinction should shape the curriculum between 2026 and 2031. The Department’s Responsible Artificial Intelligence Strategy and Implementation Pathway organizes military AI adoption around governance, warfighter trust, lifecycle management, requirements validation, an AI ecosystem and workforce development, while the Department’s five ethical principles — responsible, equitable, traceable, reliable and governable — require users to understand not only what a system produces but whether it should be trusted.

The educational standard should therefore move beyond generic AI literacy. Senior officers do not need to become machine-learning engineers, but they should be able to determine when a model is operating outside validated conditions, when outputs lack traceable provenance, when automation bias is influencing the staff, when two models disagree for material reasons, and when the appropriate command decision is to slow the process rather than accelerate it.

The urgency became visible in 2026, when Army War College faculty tested ChatGPT, Gemini, Claude and Grok against a simulated comprehensive examination. All four systems passed under the existing strategic-thinking rubric, although performance varied and degraded under sustained questioning. That result does not demonstrate that large-language models possess military judgment; it demonstrates that traditional academic assessment can be satisfied by systems that bear no command responsibility. The consequence is institutional: if a machine can generate the product being graded, the war college must increasingly evaluate the reasoning process, oral defense, adaptation and professional ownership behind the product.

Classified realism has value, but too much classification can narrow the mind

The movement toward classified planning laboratories and operationally realistic wargaming is defensible because senior officers must eventually work with real capabilities, threat assessments and vulnerabilities. The June 2026 Naval War College AI project shows the value of that environment: sensitive operational problems can be examined with data and tools unavailable in an open classroom.

But classification imposes a cost. It limits external scholarly challenge, reduces participation by some allies, narrows access to civilian expertise and can allow institutional assumptions to survive because the evidence supporting them cannot be tested broadly. The solution is not less realism but a dual architecture: classified environments for current operational problems and unclassified environments for strategic argument, historical comparison, coalition education and methodological challenge.

The Army War College’s Joint Warfighting Program, which uses an unclassified operational-level wargame spanning the 2025–2050 period, demonstrates the value of that balance. Students can work through all-domain warfare, force flow, sustainment, operational tempo and campaign sequencing without making classification the condition for every serious professional discussion. For a force that expects to fight with allies, this is not an academic preference; it is an interoperability requirement.

Reform will fail if the Pentagon measures technology instead of learning

The most serious policy error between 2026 and 2031 would be to judge PME modernization by the number of AI courses, simulations, gaming hours or secure laboratories created. CJCSI 1800.01G already provides a better model by distinguishing direct and indirect assessment and defining institutional effectiveness through evidence of learning. That framework should now be extended to the technologies entering the curriculum.

A credible scorecard would measure whether strategic content remains protected while AI use rises; whether students disclose machine assistance; whether graduates can defend conclusions without AI support; whether wargames test adaptation rather than rehearsed planning; whether faculty AI literacy rises with student use; whether exercises include corrupted and deceptive data; and whether classified modernization is matched by unclassified and multinational learning. None of these measures requires a single numerical ranking of institutional quality, but together they reveal whether reform is producing stronger judgment or merely more visible technology.

The pacing problem is faculty. Army War College analysis in March 2026 described the integration of wargaming into PME as labor-intensive and demanding for instructors; AI raises the requirement further because faculty must understand enough about models, data, security and failure modes to supervise their use. The Pentagon can purchase software faster than it can create experienced educators. That gap is likely to become the binding constraint on reform.

By 2031, the cost of getting the architecture wrong will fall on command

The next 12–24 months will determine whether the reforms initiated in 2026 become curricular integration or curricular displacement. The first indicators will be visible in academic-year course allocations, faculty recruitment, classified laboratory investment, wargaming requirements, AI-use policies and Joint Staff assessment standards. If strategy, history and campaigning remain stable while AI becomes embedded across them, the system will be moving toward integration. If technical content expands primarily by removing the time required for strategic reasoning, the Department will be trading durable judgment for rapidly depreciating proficiency.

The financial cost of inaction will appear first in educational infrastructure and workforce duplication, because the Department is already committing $13.4 billion in FY2026 to autonomy and autonomous systems while senior institutions must learn how to command forces built around those investments. The operational cost will fall on combatant commands if graduates understand neither the technologies they are expected to supervise nor the strategic limits of machine-supported decision-making. The alliance cost will fall on partners if U.S. education becomes increasingly classified and technologically inaccessible. The institutional cost will fall on civilian leaders if senior officers arrive with sophisticated tools but weaker capacity to explain what military action can achieve, what it cannot, and when faster action is strategically worse than deliberate judgment.

The choice for the Pentagon is therefore narrower than the rhetoric surrounding military transformation suggests. Between 2026 and 2031, it must modernize the instruments of education without mistaking those instruments for its purpose.

Banking System Under Fire: Why AI Is Forcing a New Cybersecurity Architecture for Global Finance

---

Navigational Index

Pillar One — The Strategic Education Function

Statutory purpose of professional military education; distinction between training and education; strategy, campaigning and joint command; institutional responsibility for producing officers capable of linking military means to national objectives; persistence of uncertainty as the central educational problem.

Pillar Two — The Technological Compression of War

Artificial intelligence, autonomous systems, drones, algorithmic decision support, multi-domain operations and digitally enabled wargaming; implications for decision speed, command structures and curriculum; advantages and limitations of technology-centred educational reform; emerging models of AI-supported learning.

Pillar Three — The 2026–2031 Education Architecture

Curricular integration rather than displacement; protected strategic foundations; experimental laboratories and classified wargaming; AI literacy and human judgment; institutional metrics; indicators of successful reform; principal risks, unresolved questions and policy pathways.

---

Europe’s Cyber Risk: Ransomware and Critical Systems

Master Abstract

The real problem is educational allocation, not technological resistance

American professional military education is entering a period in which the volume of knowledge considered operationally relevant is expanding much faster than the time available to teach it. Artificial intelligence now influences intelligence processing, logistics, targeting, decision support, modeling, planning and potentially command relationships; autonomous and semi-autonomous systems are altering force design; inexpensive unmanned systems are imposing new demands on tactical adaptation; and digitally assisted wargaming increasingly permits military institutions to expose students to complex operational problems at lower physical cost than field experimentation. The institutional case for incorporating those developments into officer education is therefore substantial rather than speculative. The Department has already formalized AI workforce education as an organizational requirement: its Responsible Artificial Intelligence implementation framework calls for standardized curricula addressing AI benefits, limitations, risk and security across military and civilian levels, while current War College initiatives are exploring AI both as an educational instrument and as a component of modern wargaming. Media Defense

Yet senior professional military education performs a function different from system qualification, technical training or operational conversion. Under 10 U.S.C. Chapter 107, Joint Professional Military Education is structured around national military strategy, joint planning at all levels of war, doctrine, command and control, force development and operational contract support; Phase II additionally includes national security strategy, theater strategy and campaigning, joint planning processes, and the integration of joint, interagency and multinational capabilities. These statutory requirements reveal the institutional logic of JPME: senior officers are educated not merely to employ military capabilities but to understand how different forms of military power interact with political authority, strategic objectives, allies, institutions and campaigns. [10 U.S.C. Chapter 107 — Professional Military Education — U.S. House Office of the Law Revision Counsel] Codice degli Stati Uniti

The distinction between training for anticipated conditions and education for unanticipated conditions consequently becomes more important as technology accelerates. Training can optimize performance against a defined problem; education must prepare officers to reformulate the problem when the environment invalidates prior assumptions. AI creates an especially revealing paradox: the technology increases the amount of information commanders can process and the speed with which options can be generated, but those improvements increase rather than eliminate the need to determine which objectives deserve pursuit, what political consequences different courses of action create, which assumptions are embedded in models, when machine-generated recommendations should be rejected and how tactical success contributes to a sustainable strategic outcome. The greater the computational capacity surrounding command, the more consequential the quality of the human judgment governing its employment becomes.

Current U.S. policy is already pushing PME toward greater warfighting relevance

The policy environment changed materially in 2026. The 6 February 2026 memorandum, “Rebuilding the Warrior Ethos in Professional Military Education,” directed a reassessment of external graduate education and explicitly framed educational investment around the requirement that graduates emerge as more capable warfighters; on 27 February 2026, the Department further described PME institutions as mechanisms for producing strategic senior leaders capable of critical thinking, multi-domain integration and service at the strategic level while restructuring Senior Service College fellowship arrangements for the 2026–2027 academic year. These documents constitute policy direction rather than evidence that existing civilian or military curricula were objectively ineffective, and their evaluative claims concerning universities should therefore be read as the Department leadership’s stated institutional position, not as independently demonstrated findings. Media Defense

The important analytical point is that the Department’s own formulation does not actually require a choice between warfighting and strategy. The February Senior Service College memorandum calls simultaneously for warfighting relevance, critical thinking, multi-domain joint integration and officers able to “serve (and think) at the strategic level,” while the legal architecture of JPME requires national security strategy and theater campaigning. The tension therefore arises primarily at the level of curriculum design and resource allocation: how much finite classroom time should be dedicated to current technological problems, how much to historical and theoretical education, and how effectively can the two be combined rather than sequentially compartmentalized. Media Defense

The 2026 National Defense Strategy further raises the pressure for operational relevance because it reorients defense priorities around present strategic requirements and a more explicit conception of military effectiveness. Its significance for education is indirect but substantial: strategic guidance changes the problems that operational institutions must study, the theaters and adversary capabilities requiring analysis, and the technologies around which force development and campaigning are organized. It does not, however, replace the statutory educational requirement to understand national security strategy, theater campaigning and joint integration. Modernization of the force therefore creates a requirement to modernize the problems studied in PME without logically requiring abandonment of the intellectual disciplines through which those problems are evaluated. Media Defense

AI can alter the educational method without replacing the educational purpose

Evidence from the Army War College demonstrates an emerging alternative to the binary model. In September 2025, War College faculty described a multi-year effort to modernize educational wargaming through AI while retaining the traditional purpose of wargames as environments in which officers develop experience through decisions, consequences and mistakes rather than merely seek a competitive “win.” On 1 October 2026, another War College publication documented the use of AI in a course spanning strategic thought from antiquity to the present, explicitly presenting AI as a tool for engaging historical strategists while preserving academic rigor, critical thinking and human judgment. War Room – U.S. Army War College

This approach has important institutional implications because it converts technology from a competitor for curricular time into a mechanism for increasing the productivity of strategic education. AI can support scenario generation, red-team argumentation, structured comparison of courses of action, rapid iteration of campaign assumptions, historical counterfactual exercises and preliminary analytical work, provided that students remain responsible for evidence, interpretation, judgment and decision. Such applications also expose students directly to AI limitations: hallucination, bias in underlying data, false precision, model dependency, opacity of reasoning, automation bias and susceptibility to manipulated or incomplete inputs become educational problems rather than abstract warnings.

Wargaming offers a similar possibility. Modern simulation can place historically grounded strategic problems inside contemporary command environments, requiring officers simultaneously to manage information, logistics, technology, coalition politics, escalation, legal constraints and political objectives. The educational objective then becomes neither nostalgia for historical campaigns nor fascination with contemporary equipment; it becomes development of transferable judgment under conditions in which tactical, technological and political variables interact.

This model is consistent with the broader JPME framework maintained by the Chairman of the Joint Chiefs of Staff. CJCSI 1800.01G, dated 15 April 2024 and still listed by the Joint Staff in 2026, provides the governing policy framework for officer PME with particular emphasis on joint education, while federal law requires the Secretary to maintain and periodically revise curricula rather than freeze them against technological change. Institutional adaptation is therefore not only compatible with the established system; revision is built into it. [CJCSI 1800.01G — Officer Professional Military Education Policy — Joint Chiefs of Staff] JCS

Lyukmanov in Cairo: Russia’s Strategic Cyber Pivot

Curricular compression is the critical vulnerability

The most important constraint is finite educational time. Every additional block devoted to autonomous warfare, software-enabled planning, AI literacy, cyber operations, space operations, information warfare or classified operational planning competes with something unless the duration, faculty base or educational delivery model changes. This means that the policy debate cannot be resolved by asserting that both technology and strategy are important; the institutional issue is which learning outcomes receive protected time and which can be delivered through integration, prerequisite education, distributed learning or professional development outside resident PME.

A technologically concentrated curriculum offers clear advantages: immediate relevance, familiarity with emerging systems, improved interaction with operational staffs and potentially greater readiness to exploit innovations appearing during the next several years. Its weakness is temporal fragility. Technology, doctrine, software and adversary countermeasures evolve rapidly, meaning specific procedural knowledge can depreciate faster than the careers of officers receiving it. Senior leaders educated principally against a prevailing operational model risk acquiring sophisticated solutions to assumptions that a future adversary, geography, coalition or technological disruption may invalidate.

A strategy-centred curriculum has the inverse profile. Historical and theoretical study does not supply detailed knowledge of contemporary sensor networks, autonomous platforms or algorithmic decision systems; without deliberate technological integration it can become detached from the conditions in which contemporary commanders operate. Its comparative advantage lies elsewhere: it develops conceptual frameworks for identifying political purpose, comparing unlike cases, challenging assumptions, understanding friction, assessing adversary agency and distinguishing operational achievement from strategic success.

The most resilient architecture consequently combines the two through strategic-technological integration: current technologies provide the problems, tools and experimental environments; strategy, history and political analysis provide the interpretive architecture; wargaming connects abstraction to decision; and assessment measures whether officers can move from tactical events through operational design to political consequence.

The requirement is strategic adaptability rather than prediction of the next war

Military education cannot reliably optimize itself around a single forecast of future warfare because senior leaders educated today may command through several different technological and geopolitical cycles. Current attention to drones, autonomous systems and AI is warranted, but the exact mix of technologies, geography, force posture, political constraints and coalition composition characterizing a future conflict remains unknowable. The educational system must therefore solve a different problem from intelligence forecasting: instead of predicting one future correctly, it must prepare officers to adapt coherently when their preferred future proves wrong.

This distinction explains why history retains operational value even when historical weapons and organizations are obsolete. The educational purpose is not imitation of historical tactics; it is exposure to a large sample of command decisions, strategic misperceptions, coalition tensions, logistics failures, political constraints, adaptation problems and interactions between military action and political objective. Contemporary conflicts provide exceptionally valuable evidence, but their very recency can obscure outcomes, causal mechanisms and second-order consequences. Historical depth expands the range of cases against which claims of technological discontinuity can be tested.

Artificial intelligence strengthens this requirement because it accelerates the production of plausible answers. Commanders increasingly require the intellectual capacity to decide whether a rapidly generated answer addresses the correct question, rests on valid assumptions and remains consistent with political objectives. In that environment, critical judgment becomes not an alternative to technical fluency but the mechanism that makes technical fluency strategically useful.

Key Evidence Table

Indicator	Value/status	Reference date	Definition/scope	Issuer	Exact source
Statutory JPME core	National Military Strategy; joint planning at all levels of war; joint doctrine; joint C2; force/requirements development; operational contract support	Law in effect Sept. 2026	Minimum statutory subject matter for JPME	U.S. Congress / U.S. Code	10 U.S.C. §2151 Codice degli Stati Uniti
JPME curriculum revision	Secretary must periodically review and revise joint PME curricula	Law in effect Sept. 2026	Department-wide statutory governance requirement	U.S. Congress / U.S. Code	10 U.S.C. §2152 Codice degli Stati Uniti
JPME Phase II	Includes national security strategy, theater strategy and campaigning, joint planning, and integration of joint/interagency/multinational capabilities	Law in effect Sept. 2026	Phase II professional military education	U.S. Congress / U.S. Code	10 U.S.C. §2155 Codice degli Stati Uniti
Current officer-PME directive	CJCSI 1800.01G remains listed as the governing officer PME instruction	15 Apr. 2024; listed 2026	Officer PME/JPME policy	Chairman, Joint Chiefs of Staff	CJCSI 1800.01G JCS
Senior leader joint education	Newly selected general and flag officers must undertake the designated course preparing them for inter-service work, subject to statutory waivers	Law in effect 29 Sept. 2026	Capstone requirement	U.S. Congress / U.S. Code	10 U.S.C. §2153 Codice degli Stati Uniti
Department PME policy shift	Department ordered reevaluation of graduate-level PME relationships and linked education more directly to warfighting effectiveness	6 Feb. 2026	Active-duty graduate PME and external programs	Secretary of War	Rebuilding the Warrior Ethos in Professional Military Education Media Defense
Senior Service College restructuring	Department directed changes to SSC fellowships while defining the intended product as strategic senior leaders capable of critical thought and multi-domain joint integration	27 Feb. 2026	2026–27 academic year onward	Secretary of War	Aligning Senior Service College Opportunities with American Values Media Defense
AI education requirement	Standardized curricula covering AI benefits, limitations, risk factors and security form part of Department AI workforce implementation	current implementation framework	Military and civilian AI workforce	Department of Defense	Responsible Artificial Intelligence Strategy and Implementation Pathway Media Defense
AI-assisted wargaming	Army War College reports a multi-year effort to use AI to strengthen educational campaign wargaming	11 Sept. 2025	Senior military education/wargaming	U.S. Army War College	Back to the Basics in Wargaming – With a Little Help from AI War Room – U.S. Army War College
Integrated AI–strategy teaching	AI used in strategic-thought education while preserving human judgment and critical thinking	1 Oct. 2026	Army War College elective instruction	U.S. Army War College	Strategic Education Powered by AI Integration War Room – U.S. Army War College

Competing Pathways for U.S. Professional Military Education

The evidence supports three genuinely different institutional pathways, although they should not be treated as equal predictions of Department policy.

Pathway	Diagnostic support	Disconfirming evidence	Indicators	Current standing
Technology-dominant rebalancing — PME reallocates substantial resident time from historical/strategic study toward AI, autonomous warfare, current operational plans and technical warfighting problems	2026 Department leadership places explicit emphasis on warfighting relevance, lethality and return on educational investment; rapid technological change increases demand for specialist competence. Media Defense	Statutory JPME requirements retain national strategy, theater strategy, campaigning and joint integration; current Army War College experimentation demonstrates that AI can be integrated without abandoning strategic education. Codice degli Stati Uniti	Reduction of strategy/history contact hours; growth in systems-specific courses; increased classified operational laboratories replacing broader seminars	Institutionally possible, but the verified record does not establish that wholesale substitution has become the governing model
Protected strategic core with technological integration — strategy, history and political-military reasoning remain foundational while AI and new technologies are incorporated into cases, planning exercises and wargames	Closely consistent with statutory JPME architecture and documented Army War College AI initiatives; allows modernization without redefining PME principally as technical training. Codice degli Stati Uniti	Requires faculty capable of crossing disciplinary boundaries, modern technical infrastructure and deliberate curriculum redesign rather than simple course substitution	Stable strategy/history requirements alongside growing AI-enabled simulations, contemporary operational cases and explicit AI-literacy outcomes	Best supported by the currently visible institutional architecture, although implementation quality cannot be established from public records alone
Expanded dual-track model — strategic education is preserved while technical content grows through longer programs, prerequisite learning, distributed education, electives or separate laboratories	Avoids direct curricular trade-off and allows technical specialization without displacing core JPME outcomes	Adds time, personnel, infrastructure, security and opportunity costs; may create duplication between operational-force training and PME	Expanded resident periods; mandatory pre-course digital modules; dedicated AI/wargaming laboratories; new faculty billets and budget lines	Attractive structurally, but public evidence does not yet establish Department-wide adoption or the resources necessary to sustain it

The second pathway currently fits the observable institutional record most closely, not because it is normatively preferable by definition, but because the statutory framework continues to protect strategy and campaigning while service institutions are demonstrating actual AI integration inside strategic education rather than replacing it altogether. That judgment should be revised if future curricula show systematic reductions in strategic study, if Department guidance formally redefines senior PME around current operational proficiency, or if new statutory language materially alters Chapter 107.

Principal Gaps and Watch Indicators

Curricular-hour allocation remains the largest unresolved evidentiary gap. Public documents establish institutional objectives but do not provide a sufficiently standardized, service-wide dataset showing how many resident hours are actually being shifted among strategy, history, operational art, technology, AI, wargaming and service-specific subjects. Without those data, claims that the system has already undergone either a major technological displacement or a major strategic renaissance would exceed the verified record.

Assessment methodology is also incomplete. Public policy documents describe desired leader attributes, but there is no common publicly documented metric demonstrating whether a graduate who completes more AI-enabled exercises subsequently performs better at linking operational decisions to strategic objectives. Educational reform should therefore be watched through outcome indicators rather than course labels alone.

Faculty composition will be diagnostic. Significant recruitment of technologists without corresponding investment in strategists, historians, political scientists, regional specialists and experienced campaign planners would indicate movement toward technical concentration; interdisciplinary appointments and jointly designed courses would indicate integration.

Classification is an important threshold variable. Increased reliance on classified simulations can improve operational realism but can also reduce engagement with civilian scholars, allies without equivalent access, publicly contestable evidence and broader academic ecosystems. The balance between classified experimentation and unclassified strategic study should therefore be monitored as an institutional design issue rather than treated merely as a security-management problem.

AI dependence itself requires measurement. The relevant question is not simply whether AI is used in the classroom, but whether officers demonstrate stronger independent reasoning after using it. Observable indicators should include the capacity to identify model errors, challenge embedded assumptions, recognize manipulated inputs, explain rejected machine recommendations and defend decisions without treating system output as authority.

The 2026–2027 academic cycle is the first significant implementation test. Department directives affecting Senior Service College opportunities and external graduate programs begin during this period, making subsequent curriculum catalogs, fellowship lists, faculty hiring, resource allocation and JPME accreditation decisions especially important evidence. Media Defense

Decision Thresholds for 2026–2031

A strategically significant change in the assessment would occur if one or more of the following became documented across the senior-service-college system: sustained reductions in national-security strategy, military history or theater-strategy requirements; formal substitution of platform or software proficiency for existing strategic learning outcomes; major expansion of classified wargaming accompanied by contraction of broad strategic education; statutory amendment to 10 U.S.C. Chapter 107 redefining minimum JPME content; or validated educational evidence showing that a different curricular model produces demonstrably superior strategic decision-making.

Conversely, the integration assessment would strengthen if the services institutionalize AI literacy as a cross-curricular competency rather than a separate technological silo, maintain protected strategy and campaigning outcomes, connect historical and theoretical study directly to contemporary operational problems, expand AI-assisted wargaming, and evaluate officers on their ability to integrate military action with political objectives rather than only on the technical quality of operational plans.

Courses of Action

Protect a statutory-strategic educational floor while permitting technological redesign

The Department can preserve the existing legal architecture of national military strategy, theater strategy, campaigning and joint integration while allowing service colleges considerable freedom over how those outcomes are achieved. The expected effect would be to prevent technological modernization from unintentionally eliminating competencies Congress already treats as intrinsic to JPME. Implementation burden would be moderate because the authority already exists within Chapter 107 and CJCS educational governance, while the principal downside would be excessive rigidity if protected subject categories are defined through hours rather than measurable competencies. Codice degli Stati Uniti

Make AI a method of strategic education as well as a subject of study

AI should be used inside planning exercises, historical analysis, scenario construction, adversarial reasoning and wargaming so that officers simultaneously learn the technology and confront its limitations. The expected effect is greater educational density: the same course can develop technological literacy, evidence discipline, strategic judgment and command decision-making. Army War College initiatives provide an existing institutional basis for this pathway. The principal risk is automation bias, particularly if assessment rewards rapid output instead of independent reasoning. War Room – U.S. Army War College

Separate perishable technical proficiency from enduring senior-leader education where appropriate

Highly system-specific knowledge can be delivered through prerequisite modules, operational assignments, specialist schools or recurring professional development, reserving scarce resident senior-college time for subjects requiring sustained seminar discussion, interdisciplinary analysis and judgment under ambiguity. The expected effect would be improved allocation of educational time, while implementation would require stronger coordination between PME institutions, operational commands and technical schools. The principal downside is fragmentation if technical knowledge becomes institutionally detached from strategic reasoning.

Build a common strategic-wargaming assessment architecture

Rather than measuring modernization through the number of AI courses or exercises conducted, institutions can evaluate whether students identify political objectives, recognize uncertainty, adapt plans after adversary action, manage coalition constraints, integrate logistics and industrial capacity, challenge model-generated recommendations and explain the strategic consequences of operational choices. Such an architecture would convert debates over curricular ideology into observable educational performance, although designing valid assessment instruments would require multi-year institutional work.

Preserve deliberate intellectual friction

A senior-officer educational environment should expose students to arguments that contradict prevailing operational concepts, because the organizational value of PME lies partly in allowing assumptions to fail before they fail in war. Historical cases, red-team exercises, competing strategic theories, AI-generated alternative courses of action and adversarial wargaming should therefore be structured around disconfirmation rather than doctrinal reinforcement alone. The principal burden is cultural: organizations optimized for operational execution do not always reward sustained challenges to dominant assumptions.

Net Assessment

The evidence available as of 1 October 2026 does not support the proposition that American professional military education must choose between preparing officers for technologically advanced contemporary war and educating them in strategy, history and political-military judgment. The statutory architecture, Joint Staff policy and emerging Army War College practice instead support a more demanding model in which the two functions are integrated.

Technology changes the instruments, tempo, information environment and organizational requirements of warfare, and a PME system that ignored AI, autonomous systems, drones, data-intensive operations and modern wargaming would progressively detach itself from the force it is intended to educate. But technological acceleration does not eliminate the problem for which senior military education exists: commanders must still decide what military action is intended to achieve, determine whether operational success advances political objectives, adapt when assumptions collapse, understand adversary agency, allocate scarce resources, manage alliances and communicate the limits of military power to civilian authority.

The principal institutional danger is therefore category error. If professional military education is treated primarily as advanced technical training, the Department risks duplicating functions already performed by operational formations and specialist schools while weakening the comparatively scarce environment in which future senior leaders can study war as a strategic and political phenomenon. If, conversely, military education treats technological change as secondary or ephemeral, it risks producing conceptually sophisticated officers unable to understand the systems and decision architectures through which contemporary military power is exercised.

The stronger architecture is one in which technology intensifies strategic education rather than displaces it: AI becomes an object of study and a tool of inquiry; wargames become laboratories for political-military judgment as well as operational planning; historical cases test claims of novelty; current conflicts supply evidence without becoming templates; and the educational objective remains the production of officers capable of exercising judgment when the future differs from the scenario for which they prepared.

---

Pillar One — The Strategic Education Function

Strategic judgment is the institutional product, not an academic accessory

The fundamental purpose of senior professional military education in the United States is not to reproduce what experienced officers have already acquired through command, exercises, deployments, staff assignments and specialist schools, but to prepare them for a qualitatively different category of responsibility in which military action must be interpreted, designed and advised as an instrument of national policy. This distinction is embedded in federal law rather than resting solely on the educational philosophy of individual war colleges. Title 10 defines Joint Professional Military Education as rigorous instruction and examination intended to produce both theoretical and practical understanding of joint matters, and identifies national military strategy, joint planning across the levels of war, joint doctrine, joint command and control, force and requirements development, and operational contract support as minimum statutory subject areas. The statutory construction is significant because it places military education inside a chain extending from national purpose through strategy and joint planning to force employment, rather than defining PME primarily around technical proficiency. [10 U.S.C. §2151 — Joint Professional Military Education] Official U.S. Code text

That architecture becomes still more explicit at senior level. The institutions identified by statute as senior-level service schools include the Army War College, College of Naval Warfare, Air War College, Marine Corps War College and Space Force Senior Level Education Program, while the Joint Staff maintains the overarching Officer Professional Military Education Policy through CJCSI 1800.01G. The result is not a collection of unrelated graduate schools but an institutional system designed to move selected officers from service expertise toward increasingly joint, strategic and policy-relevant responsibilities. [10 U.S.C. Chapter 107 — Professional Military Education] Official Chapter 107 record [CJCSI 1800.01G — Officer Professional Military Education Policy] Joint Chiefs of Staff official PME policy page

The educational function therefore begins where routine professional competence becomes insufficient. An officer commanding a formation must understand how to employ the force; an officer serving at the strategic level must additionally determine how the employment of that force interacts with national objectives, available resources, coalition commitments, institutional constraints, escalation, time, political tolerance and the actions of an intelligent adversary. Senior PME is consequently concerned with the transition from doing things correctly within an established problem to determining which problem should be solved, what outcome is politically meaningful, and which military actions are appropriate to obtain it. The National War College makes this distinction particularly visible in its mission, which centers its senior-level course of study on national security strategy and explicitly prepares graduates for the highest levels of strategic leadership in a complex and rapidly evolving environment. [National War College — Mission and Vision] National War College mission

Training and education solve different military problems

The distinction between training and education is sometimes presented too simplistically, because competent military institutions require both and because sophisticated training can involve substantial intellectual effort. The more defensible distinction concerns the structure of the problem being addressed. Training is generally optimized around performance within conditions that can be specified in advance: a procedure, system, formation, planning process, tactical problem or mission set has known standards against which performance can be measured. Strategic education increasingly operates where the relationship between action and outcome is uncertain, where the problem itself is contested, where political objectives can evolve, and where several courses of action can be technically feasible while generating profoundly different strategic consequences.

This difference explains why war colleges employ seminar discussion, analytical writing, historical cases, strategy formulation, campaign analysis, exercises and wargaming instead of relying exclusively on procedural instruction. The Marine Corps War College, for example, explicitly describes its educational design as one intended to challenge assumptions, develop alternatives and force students to articulate strategic choices, while its stated learning outcomes include evaluating national security strategies, applying an ends–ways–means–risk framework, integrating instruments of national power and assessing the advantages and disadvantages of force in pursuit of political objectives. [Marine Corps War College — Mission, Curriculum and Program Learning Outcomes] Marine Corps War College official program page

The difference can be represented institutionally rather than rhetorically:

Dimension	Training-dominant function	Senior educational function	Strategic consequence
Problem definition	Predominantly supplied	Frequently contested or incomplete	Leaders must determine what the actual strategic problem is
Desired output	Correct execution	Defensible judgment	Performance cannot always be reduced to a procedural standard
Time horizon	Immediate or mission-cycle	Campaign, theater, national and long-term	Short-term military effectiveness can generate long-term strategic costs
Knowledge structure	Procedures, doctrine, systems, techniques	Strategy, history, policy, institutions, resources, adversary behavior	Senior leaders must integrate heterogeneous evidence
Uncertainty	Reduced where possible	Treated as an inherent condition	Plans must remain adaptable after assumptions fail
Adversary	Usually incorporated into a defined scenario	Independent strategic actor capable of reframing the problem	Enemy adaptation affects political as well as operational outcomes
Assessment	Standards and demonstrated proficiency	Argument, judgment, decision quality, adaptation and consequence analysis	Several technically valid answers may have different strategic value
Failure mode	Incorrect execution	Correct execution in pursuit of an inadequate or counterproductive objective	Operational competence does not guarantee strategic success
Primary institutional location	Units, technical schools, exercises, qualification systems	War colleges, JPME institutions, senior leader programs	Complementary rather than substitutable functions

The practical implication is that the military cannot compensate for weakness in strategic education merely by improving tactical or technical training, because the two activities reduce different forms of risk. Training reduces the probability that a force will fail to perform an assigned task; strategic education is intended to reduce the probability that the institution will perform an assigned task effectively while misunderstanding its relationship to the wider objective.

The statutory architecture creates a vertical chain from force to policy

The legal structure of JPME is particularly important because its subject areas reveal what Congress expects senior military professionals to understand. National Military Strategy establishes the connection with national objectives; joint planning at all levels of war links strategic direction to operational design; joint doctrine establishes common conceptual and procedural foundations; joint command and control concerns the authority and architecture through which military action is directed; joint force and requirements development connects current operations to future capability; and operational contract support recognizes that military power depends upon resources and supporting systems extending beyond combat formations. [10 U.S.C. §2151] U.S. Code — Section 2151

This should not be read as a syllabus. It is better understood as a statutory strategic dependency chain:

Statutory domain	Strategic question generated	Failure if inadequately understood
National Military Strategy	What military contribution supports national objectives?	Force employment becomes disconnected from policy
Joint planning	How should objectives be translated into integrated military action?	Services or components optimize independently
Joint doctrine	What common conceptual framework enables joint action?	Friction, incompatible assumptions and organizational fragmentation
Joint command and control	Who decides, under what authority, with what information and coordination?	Decision latency, duplication, authority ambiguity
Force and requirements development	What capabilities must exist for future missions?	Present operational demand distorts future-force decisions
Operational contract support	What non-organic resources sustain operations?	Strategic plans underestimate logistical and commercial dependencies

The educational importance lies in the interaction among these domains. A force-development decision affects future campaign options; campaign choices create resource requirements; command arrangements determine decision authority; coalition participation changes what is politically and operationally available; contracting dependencies influence endurance; national strategy defines which of these costs can be justified. Senior PME is therefore an exercise in integration across systems whose individual optimization does not automatically produce a coherent national strategy.

Senior PME sits at the policy–strategy–operations boundary

The National War College provides the clearest institutional expression of the policy end of this spectrum. Its mission is explicitly joint, interagency and international, and its current institutional description reports that approximately 59 percent of the student body consists of representation from the land, air and sea services, including Marine Corps and Coast Guard participation, while the remaining 41 percent comes from the Department of State, other federal agencies and international fellows. That composition is educationally consequential: it deliberately places military officers inside an environment where military logic competes and interacts with diplomatic, interagency, allied and governmental perspectives. [National War College — Mission and Vision] National War College official mission page

The National War College’s current academic materials reinforce the same institutional function. Its strategy curriculum begins with strategic logic, assessment of the strategic environment, national interests, threats, opportunities, assumptions, power and the instruments available to political leaders, thereby forcing students to begin above the level of military operations rather than treating military force as the independent starting point of strategy. The College’s National Security Strategy Primer explicitly states that coherent strategy is difficult because complexity and uncertainty are intrinsic to strategic problems and that its framework is intended as a foundation for disciplined strategic reasoning rather than as a single doctrinal answer. [National War College Student Academic Handbook AY 2025–2026] National War College official website [A National Security Strategy Primer] National War College strategy resources

The Naval War College approaches the same boundary through a somewhat different institutional lens. Its College of Naval Warfare core combines Joint Maritime Operations, Theater Security Decision Making, Strategy and Warfare, Leadership in the Profession of Arms, and Perspectives on Modern War, and explicitly describes its objective as developing officers able to understand strategy and operations, think critically, confront uncertainty and surprise, operate jointly, and comprehend the broader security environment and instruments of national power. Its Strategy and Policy course concentrates directly on the relationship between purpose, objective and means and on assessing whether alternative strategies can achieve national-level objectives. [College of Naval Warfare — Core Curriculum] U.S. Naval War College core curriculum [Strategy and Policy] Strategy and Policy course

The Marine Corps War College makes the linkage still more explicit in its Joint Warfare course, which states that the course is not principally a military-planning course, but is instead intended to develop deeper understanding of how instruments of national power combine to accomplish national strategic objectives at theater level and above. Its Campaigning and Warfare course connects theory, military history, wargaming and strategic concepts to the application of force, while the program’s stated outcomes require graduates to evaluate force or the threat of force in relation to political objectives. [Marine Corps War College — Curriculum] Marine Corps War College curriculum and learning outcomes

The Air War College follows the same strategic logic through a curriculum that, for Academic Year 2026, allocates its 34-semester-hour Master of Strategic Studies structure across foundations of strategy, future air-space-cyber power, great-power studies, global campaigning, regional security, strategic leadership, contemporary strategy, a global challenge exercise, and research or electives. The distribution is analytically useful because it demonstrates that senior airpower education does not treat service-specific military capability as a self-contained field; it embeds that capability inside strategy, campaigning, regional context and leadership. [Air War College Curriculum — published 1 August 2026] Air War College official curriculum

Senior-service-school educational architecture

Institution	Explicit strategic function	Operational/campaign component	Broader policy/resource component	Distinctive institutional contribution
National War College	National security strategy and highest-level strategic leadership	Joint warfighting incorporated into strategic logic	Strong interagency, diplomatic and international representation	Integration of military power with the full national-policy environment
Army War College	Strategic-level leadership and application of landpower	Strategy, operations and responsible command	National security and resource management	Landpower within joint and national strategic problems
College of Naval Warfare	Strategic decision-making and relationship between policy, strategy and operations	Joint Maritime Operations and theater-level planning	National security decision-making and national instruments of power	Maritime power integrated into joint and national strategy
Air War College	Strategic studies and strategic leadership	Global campaigning; air, space and cyber power	Great-power and regional-security study	Air/space/cyber capabilities contextualized within strategy
Marine Corps War College	Strategic advising, military strategy and critical judgment	Campaigning, Joint Warfare and all-domain operations	National security, diplomacy, deterrence and instruments of national power	Explicit emphasis on strategic advice and political objectives
Eisenhower School	Strategic leadership through national-security resource strategy	Links resources to warfighting capacity	Industrial capacity, economic resources and resource mobilization	Connects strategy to the material ability of the state to execute it

Sources: [Army War College — MEL-1 Programs] Army War College senior programs; [National War College — Mission] National War College mission; [College of Naval Warfare — Core Curriculum] Naval War College core curriculum; [Air War College — Curriculum] Air War College curriculum; [Marine Corps War College] Marine Corps War College official page; [Eisenhower School — Mission] Eisenhower School official mission.

The strategic leader is an institutional translator

One of the least appreciated functions of senior PME is that it prepares officers to translate between communities that organize problems differently. Political leaders reason in terms of national interests, domestic legitimacy, alliances, risk and policy outcomes; military planners must translate those objectives into executable campaigns; services think partly in terms of forces, readiness and capability; combatant commands must integrate those capabilities geographically or functionally; logisticians and resource institutions determine whether the resulting concept can be sustained; allies introduce additional authorities, capabilities and political constraints. Strategic leadership therefore requires fluency across institutional languages rather than expertise in only one.

This is why the Eisenhower School occupies an important place in the PME architecture. Its formal mission is to educate joint warfighters and national-security leaders not only to analyze strategy but also to evaluate, marshal and manage the resources required to execute it. Its institutional logic rests on a proposition that strategic education must extend beyond deciding what military power should accomplish to determining whether the state possesses the economic, industrial and organizational capacity to sustain the chosen strategy. [Eisenhower School for National Security and Resource Strategy — Mission] Eisenhower School official mission

This creates a second strategic dependency chain that is often obscured when war is viewed primarily through operational performance:

Political objective → military strategy → campaign design → force requirement → industrial/resource requirement → sustainment → adaptation → political reassessment

Failure at any point can invalidate success elsewhere. A strategically elegant campaign that exceeds industrial endurance is defective; a technically superior force whose employment fractures the coalition can be strategically counterproductive; an operational success that requires resources disproportionate to the political objective can be unsustainable; a resource-efficient concept that cannot produce the required military effect remains inadequate. Senior military education exists partly to teach officers to reason across these discontinuities rather than optimize a single segment of the chain.

Campaigning is the bridge between strategy and operations

Campaigning occupies a central position because it is where abstract strategic aims encounter actual military action over time. The Naval War College’s Joint Military Operations curriculum specifically concentrates on theater-strategic and operational levels of war and asks students to develop operational concepts, apply joint planning processes, employ instruments of national power and evaluate solutions to ill-structured problems. Its capstone planning exercise then requires students to synthesize those concepts in a realistic staff environment. [Joint Military Operations — U.S. Naval War College] Joint Military Operations course

The significance of campaigning is analytical rather than merely procedural. Strategy identifies the relationship between political purposes and available power, while operations organize military actions; campaigning connects multiple operations across geography and time while constantly testing whether the underlying strategic logic remains valid. This requires senior leaders to consider sequencing, culmination, reserves, logistics, coalition endurance, escalation, force regeneration and the reactions of the adversary while simultaneously preserving alignment with political objectives.

A useful distinction is therefore:

Level	Principal question	Typical decision failure
Policy	What political condition is sought?	Objective remains ambiguous or contradictory
Strategy	How should national power create that condition?	Ends, ways and means become misaligned
Theater strategy	What military contribution and regional posture serve the strategy?	Theater action becomes detached from wider national priorities
Campaigning	How should military actions be sequenced and adapted over time?	Operational activity becomes an end in itself
Operations	How are major actions designed and coordinated?	Local efficiency obscures campaign-level consequences
Tactical action	How is the immediate military task accomplished?	Success is measured without reference to higher objectives

Senior PME is therefore most valuable at precisely those transition points where one level cannot simply be deduced mechanically from the level above it.

Jointness is fundamentally a strategic problem of integration

Joint education is sometimes reduced to knowing the capabilities and doctrine of other services, but its senior-level function is considerably broader. Jointness exists because national military problems rarely correspond neatly to service boundaries. Air, maritime, land, space, cyber and information effects interact; intelligence and logistics cut across them; allies and civilian agencies possess capabilities that military commanders do not control; and authorities can be distributed among combatant commands, service components, national institutions and coalition structures. The central problem is consequently not familiarity but integration under differentiated authorities, cultures and constraints.

The Joint Chiefs of Staff places JPME within a formal quality-management and policy framework, with the Joint Education Branch responsible for policy, programs and analysis of military education matters assigned to the Chairman. The continued listing of CJCSI 1800.01G, dated 15 April 2024, as the governing Officer Professional Military Education Policy confirms that joint educational outcomes are institutionally managed rather than left entirely to individual colleges. [Joint Chiefs of Staff — Joint Education] Joint Education official page [CJCSI 1800.01G] Joint Staff instruction listing

At senior level, effective jointness requires at least five forms of integration:

Integration problem	Senior-leader requirement	Educational implication
Service integration	Combine domain-specific capabilities without allowing service preferences to dominate the objective	Cross-service student bodies, joint planning and campaign analysis
Interagency integration	Understand instruments not controlled by military command	State Department, agency and civilian participation
Multinational integration	Reconcile national caveats, political constraints and capability differences	International fellows, coalition exercises and comparative planning
Resource integration	Connect operational ambition to logistics, industry and fiscal capacity	Resource strategy, sustainment and force-development education
Authority integration	Understand who can decide and direct different instruments	Policy, command relationships, law and civil-military decision-making

This is one reason the composition of the educational environment matters as much as the formal syllabus. A seminar containing only officers from one service can teach joint doctrine, but it cannot fully reproduce the cognitive friction created when officers, civilians, foreign partners and representatives of different bureaucratic cultures interpret the same strategic problem through different institutional responsibilities.

Strategic advice is a separate professional competency

Senior officers do not only command forces; many will advise civilian decision-makers whose questions differ fundamentally from military planning questions. A policymaker may ask not merely whether an objective can be seized, defended or destroyed, but whether doing so changes adversary behavior, whether escalation can be contained, how allies will react, what happens afterward, what resources will be required over years rather than weeks, what alternative instruments are available and what risks remain if military action succeeds exactly as planned.

The Marine Corps War College explicitly defines the development of strategic advisors as one of its graduate outcomes, while the Naval War College describes preparation for high-level policy, command and staff positions and the development of officers capable of serving as trusted advisers to policymakers. These are not incidental mission statements; they identify a professional task for which operational experience alone does not necessarily provide adequate preparation. [Marine Corps War College — Program Learning Outcomes] Marine Corps War College learning outcomes [Joint Military Operations — U.S. Naval War College] Naval War College Joint Military Operations

Strategic advice requires the officer to move beyond advocacy for the military instrument. The adviser must be able to describe what force can accomplish, what it cannot accomplish, the assumptions upon which military estimates rest, the resources and time required, potential adversary responses, possible second-order effects, and the conditions under which the original recommendation should be reconsidered. The professional standard is therefore not simply the production of a preferred course of action but the presentation of decision-quality alternatives, including their costs, uncertainty and political implications.

History functions as a database of strategic variance

At senior level, military history has value primarily because it expands the number and variety of strategic environments an officer has cognitively encountered before assuming responsibility for decisions that cannot safely be learned through personal trial and error. No serving officer can personally experience the range of coalition fractures, mobilization failures, deterrence breakdowns, resource crises, political reversals, escalation dynamics and campaign transitions that may confront a senior command. Historical study supplies indirect experience by exposing students to strategic systems with different technologies, institutions, cultures and political objectives.

The Naval War College’s Strategy and Policy curriculum illustrates this function explicitly. It combines major strategic theorists with historical case studies so that students can examine how real political and military leaders connected force to national objectives, including cases in which that connection failed. Its stated purpose is not to reproduce historical solutions but to build a framework for understanding relationships among policy, strategy and operations. [Strategy and Policy — U.S. Naval War College] Naval War College Strategy and Policy course

Marine Corps War College similarly combines theory, history and wargaming in its Campaigning and Warfare curriculum, while requiring students to evaluate both the changing character and enduring nature of war. [Marine Corps War College — Campaigning and Warfare] Marine Corps War College curriculum

The analytical advantage derives from variance. If officers study only conflicts technologically or geographically similar to the anticipated next war, they increase familiarity but reduce the diversity of strategic problems in their intellectual sample. Studying substantially different cases makes superficial imitation less useful but strengthens comparative reasoning: which causal relationships remain important despite different technologies, which assumptions were context-dependent, why apparently successful campaigns produced different political outcomes, and where decision-makers misread the relationship between force and policy.

Uncertainty is not a temporary intelligence deficiency

The persistence of uncertainty is the central reason senior military education cannot be reduced to preparing officers for a forecast operational environment. Some uncertainty can be reduced through intelligence, surveillance, modeling, exercises and better data, but a substantial portion arises from the nature of strategic interaction itself. Adversaries observe preparations and adapt; allies alter commitments; political leaders revise objectives; technologies produce unanticipated effects; logistics fail; assumptions prove false; third parties enter conflicts; domestic tolerance changes; and successful military action itself transforms the environment in which subsequent decisions must be taken.

The Naval War College explicitly lists the ability to deal with uncertainty and surprise among the outcomes of its core curriculum, while the Marine Corps War College requires graduates to reason through uncertain environments and describes its strategic-learning environment as one in which difficult decisions frequently have no uniquely correct answer. [College of Naval Warfare — Core Curriculum] College of Naval Warfare curriculum [Marine Corps War College] Marine Corps War College official page

The Army War College states the problem in similarly direct terms in its September 2026 discussion of educating for uncertainty: students arrive with extensive operational experience but are intentionally confronted with ambiguous problems, trade-offs and decisions constrained not only by force structure and doctrine but by politics and institutions. That description is consistent with the Army War College’s formal purpose of producing critical thinkers and complex problem solvers able to operate at the strategic level. [Educating for Uncertainty: What the War College Does Best — U.S. Army War College, 17 September 2026] Army War College analysis [U.S. Army War College — About] Army War College institutional mission

The principal sources of strategic uncertainty

Source of uncertainty	Why prediction is structurally limited	Educational response required
Adversary adaptation	Opponents deliberately react to observed concepts and capabilities	Competing perspectives, red teaming, adversarial wargaming
Political change	Objectives and acceptable costs can evolve during conflict	Policy analysis, civil-military reasoning, strategic reassessment
Alliance behavior	Partners possess independent governments, interests and caveats	Multinational planning and comparative political analysis
Technological interaction	Capability effectiveness depends upon countermeasures and combinations not visible in isolation	Experimentation plus historical/comparative reasoning
Operational friction	Plans encounter failures, delays, misperception and unexpected interactions	Campaign adaptation and decision exercises
Industrial endurance	Consumption and replacement rates may diverge from pre-war assumptions	Resource strategy and mobilization analysis
Escalation	Adversary thresholds are partly concealed and dynamically interpreted	Deterrence theory, signaling and strategic gaming
Information environment	Decision-makers act on incomplete and sometimes manipulated information	Evidence assessment and assumption testing
Post-success environment	Achieving an operational objective changes the political problem	Second- and third-order consequence analysis

The purpose of PME is not to eliminate these uncertainties, because no curriculum can do so. It is to produce leaders who remain intellectually functional when those uncertainties invalidate a preferred plan.

Strategy therefore requires assumption management

Every strategy contains assumptions, whether acknowledged or not: the adversary will respond in a particular way; allies will provide access; industry can replace losses; political support will endure; deterrence will hold below a threshold; a campaign will produce the expected bargaining effect; an operational advantage will remain usable long enough to matter. Weak strategic practice treats assumptions as background conditions. Strong strategic practice identifies which assumptions are load-bearing, monitors them, and establishes decision points for revising the strategy when they fail.

This explains why the National War College strategy curriculum begins with assessment of the strategic environment and the identification of assumptions rather than beginning with force employment. Its academic framework treats strategic logic as an iterative relationship among interests, threats, power, objectives and available instruments, precisely because a strategy whose assumptions cease to hold must be modified rather than simply executed more efficiently. [National War College — A National Security Strategy Primer] National War College strategy resources

This provides a useful distinction between three kinds of military competence:

Competence	Core question	What good performance looks like
Execution competence	Can the assigned action be performed effectively?	Reliable tactical or technical performance
Operational competence	Can multiple actions be organized into a coherent campaign?	Sequencing, integration, sustainment and adaptation
Strategic competence	Does the campaign remain appropriate to the political purpose as conditions change?	Reassessment of assumptions, means, objectives and consequences

The senior educational system is disproportionately responsible for the third category because it is the least amenable to repetition-based training and the most consequential when wrong.

PME is also an institutional safeguard against organizational overconfidence

Military organizations are necessarily action-oriented and hierarchical, characteristics that support execution but can produce cognitive risks when carried unmodified into strategy formation. Successful professional experience can reinforce assumptions drawn from a limited set of operational environments; service cultures can privilege particular instruments; planning processes can create an appearance of linearity; quantitative models can obscure political variables; and organizational incentives can make challenging an accepted concept professionally difficult.

The seminar structure used throughout senior PME is partly an institutional response to these risks. The Marine Corps War College deliberately employs Socratic discussion and diverse professional backgrounds to challenge assumptions, while Naval War College curricula require research papers and structured analysis precisely because strategic propositions must be defended rather than merely asserted. [Marine Corps War College — Educational Method] Marine Corps War College official page [College of Naval Warfare — Core Curriculum] Naval War College core curriculum

This does not mean academic disagreement is inherently superior to operational experience; the value lies in forcing experienced officers to expose the assumptions behind their judgments to disciplined challenge before those assumptions are embedded in real strategic decisions. PME therefore performs a governance function inside the defense institution: it creates a protected environment in which officers can examine strategic orthodoxy, test alternatives and discover conceptual failure at substantially lower cost than wartime correction.

The educational system should be assessed against strategic outputs

If the institutional purpose is strategic judgment, measurement should follow that purpose. Counting courses, reading hours, exercises or technology modules can describe inputs but does not establish educational effectiveness. A more meaningful assessment architecture would determine whether graduates demonstrate measurable improvement in the cognitive tasks they will perform at senior levels.

Decision-relevant performance framework

Strategic competency	Observable graduate behavior	Weak indicator	Strong indicator
Problem framing	Distinguishes symptoms from the underlying strategic problem	Repeats assigned problem statement	Tests whether the stated problem correctly reflects national interests
Ends–ways–means alignment	Identifies mismatches between objectives and resources	Produces internally coherent plan	Explains whether the plan can produce the political condition sought
Assumption discipline	Identifies and monitors critical assumptions	Lists generic assumptions	Distinguishes load-bearing assumptions and defines reassessment triggers
Alternative generation	Produces meaningfully different strategies	Generates minor variants of preferred option	Constructs alternatives based on different causal mechanisms
Risk analysis	Distinguishes operational risk from strategic risk	Focuses on force protection or mission failure	Includes escalation, alliance, industrial, fiscal and political consequences
Joint integration	Uses capabilities across services coherently	Adds service capabilities to a plan	Designs mutually reinforcing effects around strategic purpose
Interagency reasoning	Understands non-military instruments	Mentions diplomatic/economic tools	Explains sequencing and dependencies between instruments
Campaign adaptation	Revises operations when conditions change	Preserves original plan	Reassesses whether original strategic logic remains valid
Strategic advising	Communicates options and uncertainty clearly	Advocates one preferred course	Gives decision-makers alternatives, assumptions, costs and reversibility
Learning under uncertainty	Updates judgment from new evidence	Defends previous assessment	Changes recommendation when diagnostic evidence warrants it

The strength of such an approach is that it does not dictate a single ideological model of PME. A historically intensive course, a contemporary campaign exercise, a seminar in strategic theory or a sophisticated wargame can all contribute if they produce the relevant capability. The standard becomes whether an officer can reason better about difficult strategic decisions, not whether a curriculum conforms to a preferred disciplinary fashion.

Institutional differentiation is an advantage if the strategic core remains interoperable

Uniformity across the war colleges would be counterproductive because service perspectives contain legitimate strategic differences. Landpower, maritime strategy, air and space power, resource mobilization and expeditionary warfare present different operational and institutional problems, and the educational system benefits when each college develops expertise corresponding to the strategic contributions of its service or mission. The comparative evidence shows precisely this differentiation: Army War College emphasizes strategic landpower; Naval War College embeds maritime operations in strategy and national decision-making; Air War College combines global campaigning with air, space and cyber power; Marine Corps War College integrates campaigning with strategic advising; and the Eisenhower School concentrates on national-security resource strategy. [Army War College Programs] Army War College MEL-1 education [Naval War College] College of Naval Warfare core curriculum [Air War College] Air War College curriculum [Marine Corps War College] Marine Corps War College curriculum [Eisenhower School] Eisenhower School mission

The requirement is therefore not curricular sameness but strategic interoperability: graduates should arrive at joint and national-level assignments capable of bringing distinctive professional expertise into a common framework of policy, strategy, campaigning, resources, command and joint action. The statutory JPME system provides the minimum common architecture, while individual institutions add service and mission depth.

Key judgments

The first judgment is that senior professional military education performs an institutional function that cannot be replaced by operational training because its central object is not the execution of known military tasks but the exercise of judgment across uncertain relationships between national objectives, military strategy, campaigning, resources and political consequence. Federal law reinforces this interpretation by defining JPME around national military strategy, joint planning, command and control, force development and associated joint responsibilities. [10 U.S.C. §2151] Official U.S. Code record

The second judgment is that campaigning is the principal intellectual bridge between strategy and military action, because it is where objectives, operations, logistics, force structure, time, adversary adaptation and political constraints must be continuously reconciled. Current curricula at the Naval and Marine Corps War Colleges explicitly place theater strategy, campaigning and joint operations inside the broader problem of achieving national objectives rather than treating operational planning as an independent end. [Joint Military Operations — Naval War College] Joint Military Operations course [Marine Corps War College — Joint Warfare] Marine Corps War College curriculum

The third judgment is that uncertainty should be treated as a permanent strategic condition rather than a temporary information deficit. Intelligence and technology can narrow particular uncertainties, but interaction with adversaries, allies, political authorities and changing resource conditions continually generates new ones. The educational objective is therefore adaptability of judgment rather than perfect anticipation.

The fourth judgment is that diversity among senior PME institutions is strategically useful when it produces complementary expertise, while JPME provides the common intellectual framework necessary for joint employment. The appropriate institutional measure is not curricular uniformity but whether officers from different services can integrate their expertise into coherent national and theater strategies.

The fifth judgment is that the professional output of senior PME should ultimately be assessed through decision quality: the capacity to frame problems accurately, expose assumptions, construct alternatives, connect military means to political ends, identify second-order consequences, adapt campaigns, integrate multiple instruments of power and provide civilian leaders with clear and defensible strategic advice.

What would change the assessment

The assessment would materially change if statutory JPME requirements were altered so that national strategy, joint planning and theater-level integration ceased to constitute central educational functions; if the senior service colleges formally shifted their mission from strategic education toward principally technical or procedural qualification; or if credible longitudinal assessment demonstrated that the strategic competencies described above were being produced more effectively outside the existing war-college/JPME structure.

It would also change if public curriculum evidence demonstrated a sustained institutional separation between strategic study and operational decision-making, because that would weaken the argument that the existing system actually performs the integrative role its statutory and institutional architecture assigns to it. Conversely, evidence that graduate performance improves when strategy, campaign analysis, history, resource analysis, joint planning and practical decision exercises are deliberately integrated would strengthen the present judgment.

Open official record

The principal unresolved official-record gap is the absence of a single standardized, publicly available dataset allowing direct comparison of contact hours, assessment methods and strategic-learning outcomes across all senior-level service schools for the same academic year. Current institutional publications provide detailed curricula and mission statements, but they do not permit a rigorous service-wide calculation of how much educational time is dedicated to strategy, campaigning, history, joint planning, resource analysis, technology, exercises and independent research under fully compatible definitions.

A second gap concerns longitudinal effects. Public institutions describe graduate outcomes extensively, but publicly available evidence is much weaker on whether specific PME interventions can be causally connected to subsequent performance in senior command, joint staffs or strategic advisory positions. Any major reform claiming superior strategic effectiveness should therefore distinguish measurable educational output from institutional preference.

A third gap concerns the relationship between JPME accreditation standards and actual classroom implementation. The Joint Staff establishes the policy architecture and the institutions publish program outcomes, but deeper evaluation would require current accreditation reports, assessment evidence and comparable program-review data across the entire senior-service-school system. [Joint Chiefs of Staff — Joint Education] Joint Staff Joint Education portal

---

Pillar Two — The Technological Compression of War

Technology is compressing the interval between perception, interpretation and action

The most consequential effect of artificial intelligence on military command is not simply that computers can process information faster than human staffs, but that the entire interval separating sensing, interpretation, decision and action is being reorganized around machine-assisted processing. The Department of Defense’s data and artificial-intelligence architecture has explicitly pursued what it calls decision advantage, meaning the ability to convert data into usable understanding and military action more rapidly and reliably than an adversary. The Department’s Data, Analytics, and Artificial Intelligence Adoption Strategy identifies five principal operational outcomes: superior battlespace awareness and understanding; adaptive force planning and application; fast, precise and resilient kill chains; resilient sustainment support; and more efficient enterprise operations. DoD Data, Analytics, and Artificial Intelligence Adoption Strategy — Department of Defense (defense.gov)

The change is therefore structural. Classical headquarters collected reports, synthesized them through staffs and transmitted recommendations through command channels; contemporary architectures increasingly insert machine systems at several stages simultaneously, using automation for data ingestion, algorithms for detection and classification, predictive systems for identifying patterns, software for constructing operational pictures, and artificial intelligence for generating, comparing or prioritizing options. The result is not the disappearance of command but the redistribution of cognitive work between commanders, staffs, sensors, networks, data infrastructures and software. A major August 2026 Army War College study on AI-enabled mission command describes precisely this transformation: AI does not merely accelerate an unchanged headquarters process, but redistributes cognitive labor across people, data, tools and operational processes, thereby changing both the speed and architecture of command. Fighting with Data: Design Implications for AI-Enabled Mission-Command Systems — U.S. Army War College, August 2026 (ssi.armywarcollege.edu)

The educational consequence is fundamental. An officer no longer needs only to understand what information is available, but also how that information has been filtered, fused, ranked or generated before reaching the human decision-maker. Machine mediation therefore becomes part of command literacy. The question asked of the commander changes from “What do the reports indicate?” to “What produced this picture, what evidence did the system exclude, what assumptions structure the recommendation, how confident should the headquarters be, and what happens if the algorithm is wrong at the speed at which operations are now being conducted?”

The technological compression is occurring across several mutually reinforcing systems

Artificial intelligence, autonomy, uncrewed systems and multi-domain command should not be treated as separate technological trends because their military significance comes increasingly from their interaction. AI becomes operationally valuable when supplied with accessible and well-structured data; autonomous systems become more valuable when they can coordinate, navigate, detect or prioritize targets with reduced human intervention; distributed sensors matter when information can be fused; command networks matter when decisions can be transmitted quickly; and digitally enabled wargaming becomes more powerful when those same technologies can be represented and tested before combat.

The Department’s Combined Joint All-Domain Command and Control, or CJADC2, architecture expresses this interaction explicitly. The Department describes CJADC2 not as one platform or acquisition program but as an approach combining software applications, integrated data, operational concepts and decision-support capabilities across military domains. In February 2024 the Department announced an initial minimum viable CJADC2 capability intended to provide commanders with information and decision advantage across the force. Hicks Announces Delivery of Initial CJADC2 Capability — Department of Defense (defense.gov)

The earlier JADC2 implementation architecture explains the intended logic in operational terms: the Joint Force should be able to sense, make sense and act across the battlespace by combining automation, artificial intelligence, predictive analytics and machine learning with resilient communications and data infrastructure. The Department described the volume and complexity of modern military data as exceeding what human-intensive processes could adequately exploit, thereby making machine-supported interpretation not merely an efficiency improvement but a command requirement. DoD Announces Release of JADC2 Implementation Plan — Department of Defense, March 2022 (defense.gov)

The technological compression stack

Layer	Principal technology	Operational effect	Command implication	PME requirement
Sensing	ISR platforms, satellites, UAS, distributed sensors	Expands observable battlespace	Information volume exceeds unaided human processing	Understand sensor limits, coverage, deception and collection bias
Data infrastructure	Cloud, edge computing, data fabrics, interoperable networks	Moves data between systems and echelons	Command increasingly depends on digital infrastructure	Understand data provenance, interoperability and degraded-mode operations
Analytic layer	Machine learning, predictive analytics, computer vision	Converts raw data into patterns and alerts	Staff attention becomes algorithmically prioritized	Understand model limits, false positives, false negatives and confidence
Decision support	AI assistants, course-of-action comparison, automated planning tools	Compresses planning and assessment cycles	Human choice increasingly begins from machine-generated options	Preserve independent problem framing and alternative generation
Autonomy	Autonomous navigation, teaming, swarming, robotic control	Reduces continuous human control requirements	Command shifts toward setting constraints and intent	Understand delegated authority, control boundaries and failure modes
Effects	Networked kinetic and non-kinetic systems	Accelerates sensor-to-effect chains	Time available for verification can shrink	Develop judgment under compressed timelines
Assessment	Automated battle-damage and operational analysis	Speeds feedback into subsequent action	Action and reassessment become increasingly continuous	Teach commanders to distinguish fast assessment from valid assessment
Simulation	AI-enabled wargames, synthetic environments, digital twins	Permits rapid experimentation	Planning and education converge with simulation	Teach model criticism as well as model use

The compression does not occur equally across every mission. Strategic decisions involving political escalation, alliance relations or long-duration resource allocation remain slower and institutionally distributed, while tactical identification, targeting, navigation or logistics processes can be compressed much more aggressively. This asymmetry creates one of the central command problems of AI-enabled warfare: operational processes can accelerate faster than the political and strategic processes governing them.

Decision superiority can become decision overload

The aspiration behind AI-enabled command is usually expressed as decision advantage, but information superiority is not automatically decision superiority. A headquarters receiving more sensor feeds, more alerts, more algorithmic outputs and more potential courses of action can become cognitively slower rather than faster if the architecture transfers unfiltered complexity to humans.

The Joint Staff acknowledged this problem during development of JADC2. Senior officials described modern forces as receiving torrents of sensor information that could no longer be handled through traditional human-intensive processes, making machine learning and AI necessary to exploit data that would otherwise be lost or ignored. Joint All-Domain Command and Control Briefing — Department of Defense (defense.gov)

Yet the solution creates its own risk because AI systems can multiply outputs as easily as they reduce workload. A staff capable of generating twenty plausible options in seconds has not necessarily improved upon a staff that once generated three carefully examined options if human leaders cannot determine which assumptions, data sets and causal logic separate those alternatives. The scarce resource in AI-enabled headquarters may therefore shift from information to attention, validation and judgment.

This produces a different cognitive economy:

Traditional constraint	AI-enabled constraint
Insufficient information	Excess information of unequal quality
Slow staff synthesis	Fast machine synthesis requiring verification
Limited alternative generation	Excess alternatives requiring discrimination
Human clerical workload	Human validation and supervision workload
Delayed situational awareness	Rapid but potentially unstable situational awareness
Difficulty calculating consequences	Ease of generating modeled consequences that may conceal assumptions
Slow transmission	Fast transmission capable of propagating error quickly

The important implication for professional military education is that leaders require training not only in extracting value from AI but in refusing low-quality acceleration. Command effectiveness depends on recognizing situations in which speed is decisive and situations in which speed increases the probability of irreversible error.

Speed changes command relationships, not merely command tempo

The concept of faster decision-making is often treated as if the existing chain of command simply operates more quickly, but AI-enabled warfare can alter the organizational distribution of authority itself. When lower echelons receive fused intelligence, autonomous capabilities and machine-assisted planning tools that were historically available only at large headquarters, they acquire greater practical capacity to make complex decisions independently. Conversely, when networks permit senior commanders to observe tactical activity at unprecedented resolution, technological connectivity can encourage centralization.

This produces a command paradox: the same technology can support either decentralization or micromanagement.

The Department has repeatedly connected distributed autonomous systems with mission-command principles, particularly where forces must continue operating despite limited or disrupted communications. When introducing the Replicator initiative, senior defense leadership emphasized that attritable autonomous systems could operate in distributed environments and contribute to mission command even where bandwidth was intermittent, degraded or denied. The Urgency to Innovate — Department of Defense (defense.gov)

At the same time, the Joint Force has explicitly recognized that data connectivity cannot be assumed. Department discussions of future all-domain operations have stressed that commanders must remain capable of decentralized execution when communications are disrupted, meaning AI-enabled command cannot be designed only for permanently connected conditions. DOD Focuses on Aspirational Challenges in Future Warfighting — Department of Defense (defense.gov)

Four possible command effects of technological compression

Technological condition	Possible command effect	Advantage	Principal vulnerability
Rich shared operational picture	Greater centralized awareness	Senior command sees cross-theater relationships	Tactical micromanagement
Edge AI and autonomous systems	Greater decentralized execution	Faster local adaptation	Divergent actions and harder oversight
AI-generated recommendations	Decision delegation to software-supported staffs	Greater processing capacity	Automation bias
Network disruption	Return to mission-command autonomy	Resilience	Local decisions made with degraded strategic context

Professional education must therefore prepare officers for variable command architectures, not a single ideal system. The commander of an AI-enabled force may need to operate alternately as a centralized integrator, a decentralized intent-setter, a supervisor of machine-supported processes and a commander operating with severely degraded digital support.

Autonomous systems are moving from specialist capability toward force architecture

Autonomy is becoming materially significant because the Department is no longer treating uncrewed systems only as specialized high-value platforms. The Replicator initiative was explicitly designed to accelerate fielding of multiple thousands of all-domain attritable autonomous systems, combining lower-cost platforms with scalable production and software-enabled autonomy. Replicator’s initial tranches included uncrewed aerial systems, uncrewed surface vessels, counter-UAS capabilities and enabling software. Replicator First Tranche — Department of Defense Additional Replicator Capabilities — Department of Defense (defense.gov)

The financial trajectory confirms that autonomy has moved beyond experimental scale. In the FY2026 defense budget request, the Department identified $13.4 billion for autonomy and autonomous systems, including approximately $9.4 billion for unmanned and remotely operated aerial vehicles, $1.7 billion for surface autonomous systems, approximately $210 million for autonomous ground vehicles, and about $1.2 billion for enabling autonomy capabilities and software. The same budget briefing reported $5.3 billion in Department of the Navy autonomy investment across relevant systems. Background Briefing on FY 2026 Defense Budget — Department of Defense (defense.gov)

FY2026 autonomy request: scale of the transition

Capability category	FY2026 request/identified amount	Approximate share of stated $13.4bn autonomy portfolio*	Educational implication
Unmanned/remotely operated aerial vehicles	$9.4bn	~70%	Air-domain autonomy can no longer be treated as a niche subject
Autonomous surface systems	$1.7bn	~13%	Maritime command increasingly requires human-machine fleet integration
Autonomous ground vehicles	$210m	~2%	Ground autonomy remains smaller financially but relevant to logistics and maneuver
Autonomy software/enabling capabilities	$1.2bn	~9%	Software becomes a combat-capability layer rather than administrative support
Other components within stated autonomy total	Remaining balance	~6%	Portfolio complexity extends beyond individual platform categories

*Shares are calculated from the amounts stated in the Department’s FY2026 budget briefing and are approximate because the published verbal breakdown contains categories that are not fully exhaustive in the cited transcript. FY2026 Defense Budget Briefing — Department of Defense (defense.gov)

The scale matters educationally because future senior leaders will not merely approve individual autonomous platforms; they will make force-design, campaign, command-and-control and risk decisions involving fleets of systems with different degrees of autonomy, resilience and human supervision.

Drone warfare changes economics as well as tactics

Uncrewed systems have attracted attention because of their battlefield visibility, but their deeper strategic significance lies in the possibility of changing the relationship between cost, mass, precision and risk. The Department has described the emerging combination as “precise mass”: lower-cost autonomous or remotely controlled systems can complement exquisite platforms by increasing the number of sensors, effectors and expendable assets available to commanders. Defense leadership identified Replicator as a pathfinder for scaling this model and originally set an objective of fielding multiple thousands of attritable autonomous systems across multiple domains. The Future Character of War — Department of Defense (defense.gov)

The implication for campaign design is substantial. Traditional military planning often treats platforms as limited high-value assets whose protection shapes operational behavior; attritable systems allow commanders to accept different loss rates, saturate defenses, distribute sensing, complicate targeting and impose unfavorable cost-exchange ratios. However, inexpensive individual platforms do not automatically produce inexpensive military capability because effective large-scale autonomous operations still depend upon communications, software, data, electronic protection, logistics, production capacity, trained personnel and integration with conventional forces.

The counter-drone side illustrates the same dynamic. The Department’s FY2026 budget overview requested $3.187 billion for counter-unmanned systems, compared with $2.247 billion enacted for FY2025, an increase of $940 million, or approximately 41.8 percent year over year. FY2026 Defense Budget Request Overview — Office of the Under Secretary of Defense (Comptroller) (comptroller.defense.gov)

Counter-unmanned systems funding trajectory

Fiscal year	Department figure	Change
FY2024 actual	$2.681bn	Baseline
FY2025 enacted	$2.247bn	−$434m from FY2024
FY2026 request	$3.187bn	+$940m from FY2025
FY2025 → FY2026		+41.8% calculated increase

Source: FY2026 Defense Budget Request Overview Book. (comptroller.defense.gov)

The Department’s counter-unmanned systems strategy explicitly states that rapid proliferation of unmanned systems is changing the character of conflict and identifies commercial innovation, AI, autonomy and networking as enabling technologies allowing militaries and non-state actors with widely different resources to generate increasingly consequential capabilities. DoD Strategy for Countering Unmanned Systems — Official Fact Sheet (media.defense.gov)

For PME, the lesson is not simply “study drones.” Officers need to understand the economics of force exchange, industrial replenishment, software adaptation, electronic warfare, distributed command, counter-UAS requirements and the interaction between low-cost mass and high-value conventional systems.

Replicator demonstrates that the technology problem is organizational as much as technical

Replicator is especially significant for military education because it was designed not merely to field hardware but to change the process through which the Department identifies, acquires, scales and integrates rapidly evolving technologies. Department statements explicitly describe the initiative as a mechanism for learning how to break institutional barriers to scale and then replicate those acquisition and integration methods elsewhere. The Urgency to Innovate — Department of Defense (defense.gov)

Replicator 2 reinforces that interpretation. Rather than simply purchasing another generation of drones, the second iteration was directed toward countering small uncrewed aerial systems around critical installations and force concentrations while addressing production capacity, technology innovation, authorities, policy, open-system architecture, system integration and force structure. Secretary of Defense Memorandum: Replicator 2 Direction and Execution (media.defense.gov)

The educational requirement therefore extends beyond knowing how autonomous systems perform. Senior officers increasingly require enough technological and institutional literacy to understand:

Question	Why it belongs in senior PME
Can the system be produced at wartime scale?	Determines whether prototype performance has campaign significance
Can software be updated quickly?	Determines adaptation rate against enemy countermeasures
Is architecture open or vendor-dependent?	Affects interoperability and strategic industrial dependence
How resilient is the data link?	Determines whether nominal autonomy survives contested environments
What level of human supervision is required?	Determines manpower and command burden
Can the platform operate under degraded communications?	Determines suitability for distributed operations
What authorities govern employment?	Determines how quickly capabilities can be used
Can allies integrate with the system?	Determines coalition utility
How rapidly can losses be replaced?	Determines campaign endurance
How quickly can an adversary copy or counter it?	Determines duration of advantage

Technology-centred education becomes strategically relevant only when officers can connect system characteristics to this broader institutional architecture.

Autonomy does not eliminate human judgment; it relocates it

The Department’s governing policy for autonomous weapon systems explicitly requires autonomous and semi-autonomous systems to be designed so commanders and operators can exercise appropriate levels of human judgment over the use of force. DoD Directive 3000.09 also requires relevant systems to demonstrate appropriate performance, reliability, effectiveness and suitability under realistic conditions, and it links AI-enabled weapon development and employment to Department AI ethical principles and responsible-AI governance. DoD Directive 3000.09 — Autonomy in Weapon Systems (media.defense.gov)

This matters because the simplistic distinction between “human-controlled” and “machine-controlled” warfare is analytically inadequate. Human judgment can enter at several different points: system design, mission planning, target constraints, geographic boundaries, rules of engagement, activation, supervision, intervention and post-action review. Greater machine autonomy at one stage does not necessarily mean disappearance of human authority; it can mean that judgment moves earlier in the decision chain.

Human judgment can migrate rather than disappear

Decision stage	Traditional human role	AI/autonomy effect	New command responsibility
Mission definition	Commander sets task	Largely unchanged	Define objective precisely enough for machine-supported execution
Information analysis	Staff interprets data	Machine filters and prioritizes	Validate data and model assumptions
Option development	Staff constructs COAs	AI can generate multiple COAs	Ensure genuine alternatives rather than algorithmic variants
Platform control	Human directly operates system	Machine performs navigation or engagement functions	Define autonomy envelope and intervention conditions
Targeting	Humans identify and validate	AI assists detection/classification	Understand confidence and classification errors
Execution	Operators continuously supervise	Supervision may become intermittent	Determine acceptable delegation
Assessment	Staff compiles outcomes	AI may automate assessment	Detect misleading or incomplete feedback

This creates a curriculum requirement that is more demanding than basic AI familiarity. Officers must learn where human judgment should be concentrated, which machine functions can be safely delegated, and which decisions remain inappropriate for automation because their political, legal or escalation consequences cannot be captured by technical optimization alone.

Responsible AI becomes an operational competency

The Department’s Responsible AI framework is often interpreted as an ethics or compliance overlay, but operationally it establishes requirements that commanders and staffs must understand if AI is embedded deeply into military processes. DoD’s framework organizes implementation around governance, warfighter trust, the AI product and acquisition lifecycle, requirements validation, an AI ecosystem and the development of an AI-capable workforce. The underlying logic is that AI systems require continuing testing, evaluation, verification and validation rather than a one-time certification event. Responsible Artificial Intelligence Strategy and Implementation Pathway — Department of Defense (media.defense.gov)

DoD’s AI ethical principles define five relevant categories: responsible, equitable, traceable, reliable and governable. Traceability requires appropriate understanding of technologies, development processes, data sources and methodologies; reliability requires defined uses and continuing safety and effectiveness testing; governability includes the ability to detect unintended consequences and disengage systems exhibiting unintended behavior. Responsible AI Implementation — Department of Defense (defense.gov)

Those principles generate direct educational competencies:

Responsible-AI principle	Command question
Responsible	Who remains accountable for the recommendation or action?
Equitable	Could the model systematically misclassify particular populations, environments or data types?
Traceable	Can staff explain how the recommendation was generated and from which data?
Reliable	Is the system operating inside the conditions for which its performance was validated?
Governable	Can the commander intervene, override or disable it when behavior becomes unacceptable?

The key educational point is that responsible AI cannot remain confined to engineers, acquisition professionals and legal specialists if commanders are expected to rely upon AI-generated information during operations.

Algorithmic decision support changes the meaning of expertise

Traditional military expertise involves internalized professional knowledge that allows experienced officers to recognize patterns quickly. AI introduces a second source of pattern recognition that can outperform or out-scale humans in narrowly defined tasks while lacking broader situational understanding.

This creates three distinct relationships between officer and machine.

The first is augmentation, in which AI performs labor-intensive analytical work while the officer retains problem framing, interpretation and choice. The second is dependency, in which the human retains nominal authority but no longer possesses enough independent understanding to challenge the system. The third is automation bias, in which outputs receive excessive credibility because they are produced rapidly, quantitatively or by technologically sophisticated systems.

The Department’s own AI workforce analysis has long emphasized that effective military AI cannot be delegated entirely to a small technical elite because personnel across the force require sufficient AI and data literacy to understand system capabilities and limitations. Army War College research similarly warns that AI systems can function as “black boxes,” creating accountability and trust problems when users cannot adequately explain outputs. Trusting AI: Integrating Artificial Intelligence into the Army’s Professional Expert Knowledge — U.S. Army War College (press.armywarcollege.edu)

In senior PME, the purpose of AI literacy should consequently not be to turn every officer into a data scientist. It should create sufficient professional understanding that a commander can distinguish between:

• a model that predicts from historical correlations and a system that establishes causal explanation;
• confidence calculated inside a model and real-world uncertainty outside it;
• a technically optimized solution and a strategically appropriate one;
• high-quality output and merely fluent presentation;
• reliable performance inside the training distribution and behavior under novel or adversarial conditions.

Generative AI creates a new educational problem because it can imitate strategic competence

The arrival of large language models changes professional education more dramatically than earlier search tools because they can generate coherent arguments, strategic options, briefing materials and apparently sophisticated assessments in seconds. The Army War College tested this directly in early 2026 by administering its oral comprehensive examination to four leading commercial AI systems. Faculty panels evaluated ChatGPT, Gemini, Claude and Grok, and all four systems passed the mock examination against the War College’s strategic-thinking rubric; the study also found meaningful performance differences that generic benchmarks did not reveal. Can AI Pass the US Army War College? — Parameters, April 2026 (publications.armywarcollege.edu)

That experiment should not be interpreted as evidence that AI possesses the professional responsibility or situational understanding of a senior officer. Its educational importance lies elsewhere: if AI can produce answers that satisfy existing assessment rubrics, educational institutions must reconsider whether those rubrics adequately test uniquely human professional competence.

This creates a new hierarchy of assessment difficulty:

Traditional educational task	Generative-AI vulnerability	Stronger assessment model
Summarize doctrine	Very high	Require application to novel contradictory evidence
Produce strategic options	High	Require defense against adversarial questioning
Draft staff paper	Very high	Require provenance and oral justification
Historical comparison	High	Require source validation and explanation of analogy limits
Create briefing slides	Very high	Evaluate underlying assumptions rather than presentation quality
Develop campaign concept	Moderate to high	Introduce dynamic adversary adaptation
Exercise judgment under uncertainty	Lower if scenario is genuinely interactive	Require real-time updating as evidence changes
Accept responsibility for decision	Not transferable to machine	Preserve human accountability

The Army War College’s 2026 experiment therefore points toward dialogue-based, adversarial and adaptive assessment rather than abandonment of AI.

AI-supported learning can deepen strategic education rather than replace it

The strongest emerging model inside PME is not one in which students are forbidden from using AI, nor one in which AI produces the finished intellectual product. It is a model in which AI is deliberately inserted into the learning environment as a digital action officer, adversarial interlocutor, simulation engine, drafting assistant or analytical opponent while faculty preserve human responsibility for evidence, reasoning and judgment.

An Army War College elective documented on 1 October 2026 uses AI in the study of strategic thought from antiquity to the present, allowing students to interact with historical perspectives while explicitly preserving academic rigor, critical thinking and human judgment. Strategic Education Powered by AI Integration — U.S. Army War College, 1 October 2026 (warroom.armywarcollege.edu)

This model can be expanded considerably.

AI-supported pedagogical functions

Educational function	AI role	Human learning objective	Principal risk
Strategic dialogue	Simulate opposing argument	Improve reasoning under challenge	False authority
Historical interrogation	Generate historically framed positions	Test concepts across cases	Invented historical details
Red teaming	Produce adversarial alternatives	Challenge preferred plans	Artificial adversary lacks genuine incentives
Staff assistance	Organize large document sets	Increase analytical throughput	Loss of source verification
Scenario generation	Create evolving contingencies	Practice adaptation	Unrealistic model assumptions
COA generation	Produce multiple candidate approaches	Improve comparative analysis	Option proliferation
Wargame adjudication	Accelerate responses and scenario progression	Increase repetitions	Hidden adjudication logic
After-action analysis	Identify patterns in game outcomes	Improve feedback	Confusing model correlation with causal explanation

This approach makes AI literacy experiential rather than theoretical. Students discover model strengths and weaknesses through use rather than through generic instruction.

Wargaming is becoming the principal bridge between technological experimentation and professional education

Wargaming is especially well suited to technological compression because it can combine operational decision-making, adversary behavior, limited information, command relationships and emergent technology inside a controlled educational environment.

The Naval War College currently conducts more than 50 gaming events each year, ranging from complex multi-sided computer-assisted games to simpler seminar games, and its dedicated wargaming facility at McCarty Little Hall encompasses approximately 111,000 square feet, including a 180-seat auditorium and multiple reconfigurable gaming spaces. Wargaming Department — U.S. Naval War College (usnwc.edu)

The College also operates wargaming as both an educational and analytical function, meaning games can support student learning while simultaneously examining strategic and operational questions for Navy and Department of Defense organizations. Wargaming — U.S. Naval War College (usnwc.edu)

The scale of diffusion is also expanding beyond the U.S. student body. In November 2025, the Naval War College introduced 100 international military students from 58 countries to contemporary wargaming methods as part of development of an International Wargaming Education Program intended to build partner capacity in game design, execution and analysis. U.S. Naval War College Introduces International Students to Wargaming (usnwc.edu)

Current wargaming indicators

Indicator	Verified figure/status	Strategic significance
Naval War College gaming events	50+ annually	Wargaming is routine institutional activity, not an occasional exercise
NWC wargaming facility	111,000 sq ft	Dedicated physical infrastructure supports persistent experimentation
Main auditorium capacity	180 seats	Supports large multi-cell and plenary game activity
International wargaming seminar, Nov. 2025	100 officers	Wargaming literacy increasingly extends into allied education
Countries represented	58	Provides a vehicle for coalition learning and shared methodology
Wargaming 101, Jan. 2025	42 participants from 20 organizations	Methodology is disseminated across DoD and government
2026 AI wargaming coding project	300+ student hours	AI integration is moving from discussion toward prototype development

Sources: Naval War College Wargaming Department, DOD Leaders Complete Naval War College Course in Wargaming Design, U.S. Naval War College Students Develop AI Tool to Enhance Wargames. (usnwc.edu)

AI is already changing wargame adjudication

The Army War College provides one of the clearest demonstrations of how artificial intelligence can alter educational wargaming. During Academic Year 2026, selected seminars experimented with large language models to accelerate free-form adjudication. Instead of forcing students to devote substantial attention to rigid game mechanics, the model generated detailed responses to student decisions, allowing seminars to concentrate more directly on operational planning and campaign design. Faculty reported positive results while also acknowledging continuing limitations. Back to the Basics in Wargaming – With a Little Help from AI — U.S. Army War College (warroom.armywarcollege.edu)

This is an important pedagogical development because one historical weakness of educational wargaming has been the trade-off between realism and usability. Highly detailed rule systems can model more variables but consume instructional time, while simple games facilitate participation but may abstract away critical operational relationships. AI offers the possibility of replacing portions of mechanical rule execution with rapid context-sensitive adjudication.

The advantage, however, depends upon auditability. If participants cannot understand why an AI adjudicator produced an outcome, the game can teach a lesson that derives from the model rather than from defensible military logic.

AI-supported adjudication should therefore distinguish at least four categories:

Adjudication type	Appropriate AI role	Required human control
Administrative events	High automation	Minimal supervision
Environmental responses	AI-supported generation	Faculty validation
Adversary operational response	AI proposes options	Human red cell retains authority
Strategic/political consequence	Limited AI support	Human adjudication should dominate

This preserves speed where machine assistance has high value while preventing the model from quietly becoming the strategic authority governing the exercise.

Naval War College experimentation is moving from AI discussion toward AI-built tools

The Naval War College provides another significant indicator. In June 2026, two student officers working through the Center for Naval Warfare Studies were recognized after spending more than 300 hours developing classified AI-enabled software intended to enhance wargaming and combat assessment. According to the College, the tool used statistical methods to improve repeatability and replicability in examining alternative military courses of action against relative capabilities and limitations. U.S. Naval War College Students Develop AI Tool to Enhance Wargames, Assess Combat — June 2026 (usnwc.edu)

The importance of this project is institutional rather than technological. Students were not merely learning about AI as an external technological trend; they were producing an analytical tool inside the professional military education and research environment. This represents a different model of war college: not only a place where officers consume established professional knowledge, but a laboratory where they participate in building and testing the analytical infrastructure through which future operational decisions may be made.

The College reinforced that trajectory in July 2026 through student and faculty AI workshops covering cybersecurity, law, adaptive learning, course design, security and wargaming applications. U.S. Naval War College Students and Faculty Discuss Uses, Potential of Artificial Intelligence — July 2026 (usnwc.edu)

The most important educational shift is from AI awareness to AI supervision

Early technology education often focuses on terminology: machine learning, neural networks, autonomous systems, generative models and related concepts. That level is no longer sufficient for senior leaders. The more important capability is AI supervision: knowing what tasks to assign, what evidence to require, what outputs deserve skepticism, when a model is operating outside validated conditions, and when a human decision-maker must slow the process down.

A mature PME curriculum should therefore separate four competence levels.

Competence level	Requirement	Appropriate population
AI awareness	Understand core concepts and terminology	Entire force
AI literacy	Understand strengths, limits, data dependence and common failure modes	Officers and staffs
AI operational supervision	Evaluate outputs, govern employment and identify when intervention is required	Commanders and senior staff
AI engineering expertise	Design, train, test and technically validate systems	Specialist technical workforce

Senior PME should concentrate primarily on the third level while ensuring sufficient mastery of the second.

Technology-centred education has genuine advantages

A stronger technological component in professional military education can produce several significant benefits.

First, it can reduce the gap between senior leadership and the technical systems through which modern forces operate. Leaders who do not understand AI, software architecture, autonomy or data infrastructure risk becoming dependent upon specialists and vendors for judgments that carry strategic consequences.

Second, technology-focused education can improve acquisition and force-design decisions because officers learn to distinguish operationally useful capability from technically impressive demonstration.

Third, digital simulations can dramatically increase the number of decision repetitions available to students. Traditional exercises are expensive and slow; synthetic environments can expose students repeatedly to alternative campaign conditions.

Fourth, AI can reduce low-value staff work and redirect time toward higher-order analysis.

Fifth, technology education can strengthen dialogue between military institutions and the commercial sector, where a substantial share of modern AI innovation originates.

Principal advantages

Advantage	Mechanism	Expected educational effect
Greater technical fluency	Exposure to real AI/autonomy systems	Senior leaders communicate more effectively with technical staffs
Faster experimentation	Digital simulation and AI adjudication	More decision repetitions
Improved acquisition judgment	Understanding software and lifecycle constraints	Better translation of operational demand into requirements
Better staff productivity	AI assists synthesis and drafting	More time for analysis
Stronger adaptation culture	Continuous experimentation	Reduced attachment to static doctrine
Improved human-machine teaming	Repeated supervised use	Better understanding of delegation limits

Technology-centred reform also creates distinctive failure modes

The fact that technological competence is increasingly necessary does not establish that more technology instruction always produces better senior leaders. The principal danger is curricular capture, in which educational reform becomes organized around the technologies receiving the greatest current institutional attention.

Technology-centred education has at least seven important limitations.

Technical knowledge depreciates rapidly

Specific software, interfaces, platforms and commercial models can change substantially within a few years, while senior officers educated today may serve at strategic level for much longer. Curriculum concentrated on current tools therefore carries high depreciation risk.

Current wars are not complete models of future war

Drones, electronic warfare and AI observed in contemporary conflicts provide essential evidence, but future wars may occur under different geography, alliance structures, escalation conditions, air superiority, maritime access or nuclear constraints. A curriculum built too narrowly around recent battlefields can convert observation into doctrine prematurely.

Faster decisions are not automatically better decisions

Machine systems can accelerate a poor planning process as effectively as a good one.

AI can create false confidence through linguistic fluency

Generative systems often present weak reasoning in authoritative form, creating a risk that polished output is mistaken for valid analysis.

Classification can narrow intellectual diversity

Technology-rich operational exercises may require classified networks and scenarios, improving realism while reducing participation by outside scholars, some allies and institutional critics.

Vendor dependence can enter the curriculum

If educational institutions rely heavily on proprietary systems, commercial architectures can shape what students understand as technically possible.

Technical focus can obscure political consequence

Algorithms optimize measurable objectives. Strategy often depends upon objectives that are contested, changing or politically qualitative.

Technology-centred reform: benefit–risk balance

Reform	Immediate benefit	Structural risk	Mitigation
More AI courses	Increases literacy	Crowds out broader subjects	Integrate AI across existing courses
More classified simulation	Greater operational realism	Intellectual closure	Pair with unclassified analytical work
More commercial tools	Faster access to advanced models	Vendor dependency	Use multiple models and open standards
More automated assessment	Faster feedback	Hidden assumptions	Require explainable adjudication
More technical faculty	Greater expertise	Disciplinary imbalance	Build mixed faculty teams
More operationally current content	Immediate relevance	Overlearning present conflicts	Use historical and alternative scenarios
More digital planning	Faster iteration	Automation bias	Require manual challenge and red-team review

The correct curriculum architecture is layered rather than technology-dominant

The strongest response to technological compression is not to create a separate technological curriculum competing with strategy, campaigning and joint command. It is to embed technological understanding into the decisions senior officers already need to make.

A strategy seminar can require AI-assisted analysis but evaluate whether students detect poor assumptions. A campaign course can include autonomous mass and contested networks as operational variables. A resource-strategy course can examine compute, semiconductor, software and industrial dependencies. A command course can examine delegated machine authority and mission command. A wargame can use AI for scenario adaptation while retaining human adjudication for politically sensitive decisions.

This produces a layered architecture:

Educational layer	Content	AI/autonomy integration
Strategic environment	Actors, interests, threats	AI competition, data infrastructure, commercial technology
Strategy formulation	Ends, ways, means, risk	AI-supported alternatives and assumption testing
Campaigning	Sequencing and operational design	Autonomous mass, decision-support systems, contested networks
Command	Authorities and mission command	Human-machine delegation and cognitive distribution
Joint integration	Cross-domain coordination	CJADC2 and machine-enabled data fusion
Resources	Industry and sustainment	Compute, software, autonomy production, data and talent
Wargaming	Decision practice	AI adjudication and synthetic environments
Assessment	Strategic judgment	Oral defense, provenance and model criticism

The effect is to make technology part of professional judgment rather than a substitute for it.

The most valuable future war college may be part school, part laboratory

The emerging evidence from Army and Naval War College initiatives suggests that the institutional model itself is beginning to change. The war college can increasingly function simultaneously as an educational institution, experimentation environment, technology testbed and analytical laboratory.

The Army War College’s operational wargaming initiatives already expose students to all-domain warfare problems extending through the 2025–2050 period, requiring decisions involving force flow, sustainment, cross-domain capabilities and operational tempo. Joint Warfighting at the U.S. Army War College (warroom.armywarcollege.edu)

The Naval War College combines academic education, operational research and recurring wargaming for external commands, while recent AI initiatives show students themselves becoming developers of analytical tools. U.S. Naval War College Wargaming (usnwc.edu)

Air War College is likewise explicitly connecting leader development, research, doctrine and wargaming to real operational problems and defines its senior educational objective around joint all-domain integration and strategic decision-making. Air War College Connects Leader Development to Real-World Decision-Making — May 2026 (airuniversity.af.edu)

This emerging model produces a continuous cycle:

Education → experimentation → wargaming → operational observation → analytical revision → curriculum update

The cycle is potentially much faster than traditional academic curriculum revision, but it should not become so fast that short-term operational observations are converted immediately into institutional doctrine.

The compression of war creates a new premium on knowing when to slow down

This is the paradox at the center of technological military education. AI, autonomous systems and integrated networks are valuable precisely because they can accelerate military processes, yet the senior commander’s distinctive function often becomes determining which decisions should not be accelerated.

A tactical sensor-to-shooter process may derive advantage from seconds saved. A decision with escalation consequences may require deliberate human review. Automated logistics routing can often tolerate extensive machine delegation. A strategic choice involving entry into a conflict, expansion of targeting categories or acceptance of coalition risk cannot be reduced to computational optimization.

The curriculum should therefore teach variable decision tempo.

Decision tempo should depend on consequence

Decision category	Appropriate tempo	AI role	Human role
Routine logistics optimization	Very rapid	High	Supervision
Sensor data triage	Very rapid	High	Validation
Defensive reaction	Rapid	High but bounded	Rules and override
Operational targeting	Rapid but controlled	Decision support	Authorization and context
Campaign redesign	Deliberate	Analytical support	Synthesis and judgment
Alliance decision	Deliberate	Scenario analysis	Political-military interpretation
Escalation decision	Highly deliberate	Limited support	Human authority dominant
Strategic objective revision	Highly deliberate	Exploratory analysis	Political leadership and senior advice

The technologically competent commander is therefore not the commander who always decides fastest. It is the commander who can distinguish where speed creates advantage from where speed destroys the time needed for strategic judgment.

The new command competency is cognitive orchestration

AI-enabled command increasingly requires senior officers to manage not only people and organizations but also distributed cognition. The headquarters becomes a combined cognitive system in which humans, algorithms, sensors, databases and software each perform different analytical tasks.

The Army War College’s 2026 work on AI-enabled mission command emphasizes this shift by describing military decision-making as a network of human and machine cognition rather than a simple relationship between commander and tool. Fighting with Data — U.S. Army War College (ssi.armywarcollege.edu)

Future senior officers therefore require the ability to answer questions that previously belonged mainly to technical specialists:

• Which decisions should machines prepare but humans make?
• Which functions can operate autonomously?
• Which data sources deserve trust?
• Who validates model performance?
• What happens when models disagree?
• What happens when communications fail?
• What information must remain intelligible to humans?
• What constitutes sufficient confidence for action?
• How should the commander detect automation bias inside a staff?
• How should authority be distributed when AI allows much faster local decision-making?

This is not software engineering. It is command theory for an AI-enabled force.

Key judgments

The first judgment is that technological compression is altering the architecture of command more deeply than the mere introduction of new platforms. AI, data infrastructure, autonomy and networked sensors collectively redistribute cognitive work among humans and machines, shorten some decision cycles and create new dependencies on software, data provenance and digital infrastructure. DoD Data, Analytics and Artificial Intelligence Adoption Strategy Fighting with Data — U.S. Army War College (defense.gov)

The second judgment is that autonomy has moved beyond an experimental phenomenon into force-structure and budgetary significance. The FY2026 defense budget request identified $13.4 billion for autonomy and autonomous systems, while counter-unmanned-system funding was requested at $3.187 billion, approximately 41.8 percent above FY2025 enacted funding. FY2026 Defense Budget Briefing FY2026 Defense Budget Request Overview (defense.gov)

The third judgment is that the critical professional competency is shifting from AI awareness toward AI supervision. Senior officers do not need to become machine-learning engineers, but they must understand enough about model limitations, data dependencies, delegated authority, testing, reliability and governability to exercise command responsibility over AI-enabled systems. Responsible Artificial Intelligence Strategy and Implementation Pathway (media.defense.gov)

The fourth judgment is that wargaming is becoming the most important institutional bridge between emerging technology and senior professional education because it permits officers to experiment with AI, autonomy, command structures and multi-domain operations without converting unvalidated technological assumptions directly into doctrine. The Naval War College now conducts more than 50 wargaming events per year, while Army War College experimentation is incorporating large language models into adjudication and campaign learning. Naval War College Wargaming Department Back to the Basics in Wargaming – With a Little Help from AI (usnwc.edu)

The fifth judgment is that generative AI is forcing PME institutions to reconsider assessment itself. If commercial AI systems can satisfy traditional strategic-writing or oral-response rubrics, the educational system must assess source discipline, adaptation, adversarial reasoning, responsibility and human judgment rather than merely polished output. Can AI Pass the US Army War College? (publications.armywarcollege.edu)

The sixth judgment is that technology-centred reform becomes counterproductive when technological recency becomes the organizing principle of senior education. Specific platforms and software depreciate rapidly, while the professional requirement is to prepare leaders capable of supervising technological systems whose precise form cannot be predicted. The more durable educational objective is therefore not mastery of today’s AI stack but the capacity to evaluate, experiment with, govern and adapt to successive technological generations.

The seventh judgment is that the decisive future command skill is cognitive orchestration: deciding what humans should do, what machines should do, how their outputs should be combined, when autonomy is appropriate, how degraded networks alter authority and when the speed of an AI-enabled system should deliberately be interrupted by human judgment.

What would change the assessment

The assessment would strengthen materially if the Department demonstrates that CJADC2, AI-enabled decision support and autonomous systems are becoming routine rather than exceptional elements of operational command; if war colleges expand measurable use of AI-supported campaign experimentation; if senior-service curricula establish formal learning outcomes for AI supervision and human-machine command; and if assessment systems begin explicitly measuring commanders’ ability to challenge or override machine-generated recommendations.

It would weaken if AI integration remains concentrated in isolated electives and research programs without changing mainstream command education, if operational systems remain too fragmented for machine-supported integration at scale, or if classified testing demonstrates that current AI systems cannot remain sufficiently reliable under contested and deceptive military conditions.

A particularly important signpost will be whether future exercises evaluate decision quality under degraded or corrupted data, rather than allowing students to experience AI only under favorable information conditions. Another will be whether senior leaders are required to demonstrate independent reasoning after AI assistance is removed.

Open official record

The public record remains incomplete regarding the exact operational performance of CJADC2 because significant portions of the implementation architecture, capabilities, vulnerabilities and testing remain classified. The Department has publicly confirmed delivery of an initial minimum viable capability, but the public record does not support a detailed assessment of wartime reliability, interoperability or resilience against a peer adversary. Initial CJADC2 Capability — Department of Defense (defense.gov)

The public record is also insufficient to measure the operational effectiveness of Replicator systems at force-wide scale. Official announcements establish selected capabilities, program objectives and acquisition progress, but not a comprehensive, independently auditable dataset of fielded inventories, readiness, mission availability, attrition, software reliability or performance under sustained electronic attack. Replicator Capability Announcement — Department of Defense (defense.gov)

A third gap concerns educational outcomes. Army and Naval War College experimentation establishes that AI is entering strategic education and wargaming, but there is not yet a publicly available cross-service dataset showing whether AI-supported instruction produces measurable improvement in strategic judgment, operational adaptation or later command performance compared with conventional educational methods.

A fourth gap concerns human-machine command at scale. Existing policy establishes requirements for appropriate human judgment and responsible AI, but publicly available records do not yet establish a mature common doctrine specifying how command responsibility should be distributed when many autonomous systems, decision-support models and human staffs operate simultaneously across multiple domains.

A fifth gap concerns the effects of adversarial AI. Most public educational discussion concentrates on exploiting U.S. AI capabilities, while future conflict will involve adversaries attempting to corrupt training data, manipulate sensor inputs, exploit model weaknesses, generate deceptive information, attack network infrastructure and intentionally induce automation errors. Senior military education will be incomplete until machine deception and algorithmic contest become routine features of professional exercises rather than exceptional technical scenarios.

---

Pillar Three — The 2026–2031 Education Architecture

Reform should change the architecture of learning, not simply add technological content

The central educational challenge for the 2026–2031 period is no longer whether artificial intelligence, autonomous systems, digitally enabled planning and modern wargaming should enter professional military education, because that transition is already under way across U.S. senior military institutions. The more consequential question is how those technologies should be incorporated without allowing rapidly depreciating technical knowledge to displace the strategic, joint, institutional and political competencies that define senior military education. Current Joint Staff policy already provides a governance framework for this problem because CJCSI 1800.01G requires systematic institutional assessment, defines professional military education in terms of learning outcomes rather than mere course attendance, and treats institutional effectiveness as an ongoing process in which evidence of student learning should influence decisions and resource allocation. CJCSI 1800.01G — Officer Professional Military Education Policy

The emerging architecture should therefore be built around a simple institutional rule: protect enduring strategic competencies, integrate technological capability into the exercise of those competencies, and assess whether the combination improves decision quality under uncertainty. This is a more demanding model than either preserving the existing curriculum unchanged or replacing strategic studies with courses on emerging technology, because it requires faculties to redesign how officers learn, how exercises are structured, how AI is used, how wargames are adjudicated, how classified material is introduced and how graduates are evaluated.

The policy environment is pushing in precisely this direction, although through several distinct authorities that should not be conflated. The 2026 National Defense Strategy emphasizes rebuilding the Joint Force, strengthening deterrence, increasing allied burden-sharing and rebuilding the defense industrial base, thereby creating strong demand for senior leaders able to understand advanced warfighting capabilities while integrating them with national strategy, industrial capacity and alliance structures. 2026 National Defense Strategy

Separately, the 2025 America’s AI Action Plan explicitly called for Senior Military Colleges to become hubs of AI research, development and talent building, including instruction in core AI skills, AI literacy, development and infrastructure management. That direction was reinforced in June 2026 by a National Security Presidential Memorandum framework directing the national security enterprise to accelerate AI adoption while maintaining systems that are robust, controllable and subject to clear chains of accountability. These documents create a strong external policy requirement for AI literacy, but neither requires the abandonment of strategic studies; the architectural problem is therefore one of integration. America’s AI Action Plan

The protected strategic core should be defined by competencies, not by preserving legacy course titles

A protected curriculum does not mean freezing a particular syllabus indefinitely. It means identifying the competencies that should remain non-negotiable even as technologies, cases and teaching methods change. The most durable way to protect strategic education is therefore not to mandate that every institution preserve an identical number of hours under historically familiar course labels, but to establish outcomes that cannot be removed when curriculum becomes crowded.

The current Joint Staff policy already supports that approach through its emphasis on learning areas, institutional assessment and continuous quality management rather than static course structures. The Air War College’s current 34-semester-hour Master of Strategic Studies curriculum illustrates how a modern institution can preserve broad strategic foundations while retaining curricular flexibility: six semester hours are allocated to Foundations of Strategy, five to Great Power Studies, five to Regional Security Studies, four to Global Campaigning, four to Air, Space and Cyber Power in the Future, two to Strategic Leadership, two to Contemporary Strategy, one to the Global Challenge Exercise, one to introductory strategic studies and four to research or electives. Air War College — AY26 Curriculum

Current Air War College degree structure as an example of balanced allocation

AY26 requirement	Semester hours	Share of 34-hour degree	Architectural function
Foundations of Strategy	6	17.6%	Protected strategic logic
Great Power Studies	5	14.7%	Adversary and systemic context
Regional Security Studies	5	14.7%	Geographic and political context
Air, Space and Cyber Power in the Future	4	11.8%	Technological and domain adaptation
Global Campaigning	4	11.8%	Operational-strategic integration
Research/Electives	4	11.8%	Adaptability and specialization
Strategic Leadership	2	5.9%	Command and institutional leadership
Contemporary Strategy	2	5.9%	Current strategic problems
Introduction to Strategic Studies	1	2.9%	Conceptual framing
Global Challenge Exercise	1	2.9%	Experiential synthesis
Total	34	100%	Integrated strategic degree

Calculated from the official curriculum published by Air War College on 1 August 2026.

This distribution is analytically important because it demonstrates that technological content can be introduced without allowing it to become the sole organizing principle of the curriculum. Only a limited share of the degree is explicitly organized around future air, space and cyber power, while strategy, regional analysis, campaigning, leadership and research remain structurally embedded.

The protected core for 2026–2031 should therefore be framed around seven enduring competencies rather than protected bureaucratic course labels.

Protected competency	Why it should be preserved	Technology integration should occur through
Strategic problem framing	Prevents technical solutions from defining the problem	AI-assisted scenario exploration and structured assumption testing
Ends–ways–means–risk reasoning	Keeps military activity tied to political purpose	Algorithm-supported option generation followed by human comparison
Campaign design	Connects strategic objectives to sequenced action	Autonomous systems, data integration and contested-network scenarios
Joint and multinational integration	Reflects actual command environment	Multi-domain simulations and coalition wargames
Strategic resource reasoning	Links plans to industry, logistics and sustainment	Modeling of compute, munitions, autonomy and supply-chain constraints
Command judgment	Determines when to delegate, intervene or slow decision tempo	Human-machine teaming exercises
Historical/comparative reasoning	Tests claims of novelty and inevitability	AI-supported comparative case analysis with source verification

The architecture should allow the content inside these competencies to change rapidly. Strategic problem framing can incorporate AI-enabled deception; campaigning can include autonomous mass; resource analysis can include semiconductor and compute dependencies; command studies can include model governance and machine delegation. What should remain protected is the intellectual function.

The 2026 policy environment increases pressure for curriculum relevance but does not eliminate the strategic requirement

Two February 2026 memoranda materially altered the policy environment surrounding senior military education. The 6 February memorandum on professional military education ordered a reassessment of external graduate education and required components to evaluate whether programs produced cost-effective strategic education for future senior leaders, while the 27 February Senior Service College memorandum described PME institutions as essential mechanisms for producing critical thinkers able to integrate multi-domain joint operations and operate at the strategic level. Those memoranda contain strongly evaluative claims about particular civilian institutions that represent Department leadership positions rather than independently established educational findings; their relevance here lies in the institutional direction they impose on the Department, namely greater emphasis on warfighting relevance, strategic-level thinking and demonstrable return from senior education. Rebuilding the Warrior Ethos in Professional Military Education — 6 February 2026

That pressure should lead to better measurement of educational outputs rather than a simplistic reduction in strategic breadth. A curriculum can become more operationally relevant by replacing generic discussion with realistic decision problems, integrating emerging technologies into strategy exercises, requiring students to work with current planning constraints and increasing experiential learning. None of those measures requires replacing strategic education with technical familiarization.

The most resilient architecture is a four-layer model

The 2026–2031 PME system should be organized as four interacting layers rather than a binary choice between “traditional education” and “technology education.”

Layer One — Protected strategic foundations

This layer should contain the competencies that must remain valid even when current technologies become obsolete: strategy, political purpose, historical reasoning, civil-military relations, joint command, campaign design, resource mobilization, institutional analysis, alliance management and strategic communication.

Layer Two — Technology integration

Artificial intelligence, autonomy, data systems, cyber, space and advanced sensing should be inserted across the strategic curriculum rather than isolated inside a small number of optional technical electives.

Layer Three — Experimental application

Students should test concepts through wargames, simulation, classified planning problems, red-team exercises, AI-assisted analysis and iterative campaign design.

Layer Four — Assessment and feedback

Institutions should measure not only whether students understand the technologies but whether those technologies improve or degrade problem framing, strategic reasoning, adaptability and command judgment.

Proposed 2026–2031 architecture

Layer	Principal purpose	Primary methods	Failure to avoid
Protected foundations	Preserve enduring strategic reasoning	Seminar, historical cases, policy analysis, theory	Curriculum becoming technologically fashionable but strategically shallow
Technology integration	Make emerging capabilities operationally intelligible	Embedded AI use, technical literacy modules, contemporary cases	Technology isolated from strategic decision-making
Experimental application	Convert knowledge into performance	Wargaming, simulation, red teams, classified labs	Academic knowledge never tested against friction
Assessment/feedback	Determine whether reform works	Direct assessment, oral defense, longitudinal tracking	Measuring modernization by course count rather than decision quality

This four-layer structure is consistent with current Joint Staff quality-management principles, which define institutional effectiveness in terms of systematic assessment, evidence of student learning and informed resource allocation.

The future educational laboratory should combine classified and unclassified environments

Senior PME requires both classified and unclassified experimentation, but they serve different institutional purposes.

Classified environments provide access to current plans, threat capabilities, intelligence assessments, system performance, vulnerabilities and operational constraints that cannot be reproduced openly. They are therefore essential for preparing officers to work inside actual planning architectures. The Naval War College’s June 2026 AI project demonstrates how significant this environment can become: two students working within the Center for Naval Warfare Studies spent more than 300 hours developing classified AI-enabled software using statistical tools intended to enhance wargaming and combat assessment through more repeatable and replicable analysis of military courses of action. U.S. Naval War College Students Develop AI Tool to Enhance Wargames, Assess Combat

The advantage is obvious: students can work with contemporary operational problems rather than sanitized abstractions. The limitation is equally important: classified environments reduce the pool of external participants, inhibit public scholarly challenge and make institutional assumptions harder to contest outside the military system.

Unclassified experimentation provides the counterweight. The Army War College’s Joint Warfighting Program uses an unclassified operational-level wargame focused on all-domain warfare across the 2025–2050 timeframe, allowing students to experiment with force flow, sustainment, domain integration, operational tempo and sequencing without requiring every learning interaction to occur inside a classified system. Joint Warfighting at the U.S. Army War College

The strongest institutional model is therefore not maximum classification but functional separation.

Educational requirement	Best environment	Reason
Current operational plans	Classified	Requires authentic contemporary detail
Adversary intelligence	Classified where necessary	Protects sources, methods and sensitive estimates
Strategic theory	Unclassified	Benefits from broad debate
Historical comparison	Unclassified	Supports academic challenge and external scholarship
Alliance planning	Mixed	Some operational detail classified, political analysis often open
AI tool experimentation	Mixed	Sensitive datasets may require closed environments; methods should be contestable where possible
Wargame adjudication	Mixed	Classified inputs can improve realism; transparent rules improve learning
Strategic writing	Primarily unclassified	Forces reasoning to survive without reliance on privileged information
Operational design exercises	Mixed	Supports realism while preserving analytical flexibility

The institutional principle should be that classification protects necessary information, not assumptions. If a proposition is taught only inside a classified environment, faculty must ensure that classification is not inadvertently preventing intellectual challenge.

Wargaming should become a permanent assessment mechanism, not an occasional pedagogical event

Wargaming is already institutionally mature enough to support a larger role in assessment. The Naval War College conducts more than 50 gaming events per year, ranging from sophisticated computer-assisted multi-sided games to seminar games, which demonstrates that experiential decision exercises can operate at significant scale rather than remaining isolated capstones. U.S. Naval War College — Wargaming

The Army War College’s 2026 analysis similarly argues that wargaming develops skills that conventional writing cannot fully test because an adaptive opponent exposes whether apparently coherent strategic reasoning survives interaction, uncertainty and consequences. When Strategy Leaves the Page — Why Wargaming Still Matters

This suggests a fundamental architectural change: wargaming should move from being principally a teaching device to becoming part of the measurement system for strategic competence.

A written strategy paper can reveal reasoning quality, evidence use and conceptual coherence. A wargame can test whether the student adapts when assumptions fail, adversaries behave unexpectedly, coalition partners refuse support or resources become unavailable.

Proposed assessment portfolio

Assessment format	Capability tested	Weight should capture
Analytical writing	Logic, evidence, strategic coherence	Can the officer explain a defensible argument?
Oral defense	Intellectual ownership	Can the officer defend and revise the argument without relying on prepared prose?
Adversarial seminar	Critical thinking	Can the officer respond to competing frameworks?
Wargame	Adaptation under interaction	Can the strategy survive adversary agency?
AI-assisted assignment	Machine supervision	Can the officer exploit AI without surrendering judgment?
Classified exercise	Operational relevance	Can the officer apply reasoning to current force problems?
Unclassified capstone	Transferability	Can the officer reason without dependence on privileged data?

A mature evaluation regime would not replace writing with gaming; it would require performance across all seven categories.

AI literacy should be standardized into four distinct levels

The current policy direction calls explicitly for AI literacy. America’s AI Action Plan states that Senior Military Colleges should become hubs of AI research and talent development and should teach core AI skills and literacy. The risk is that “AI literacy” becomes an undefined slogan producing inconsistent curricula across institutions.

A common competency model should therefore differentiate between levels.

Proposed military AI competency architecture

Level	Population	Minimum capability	PME implication
Level A — Familiarity	All personnel	Recognize basic AI concepts and common uses	Delivered before senior PME
Level B — Professional literacy	Officers and staffs	Understand data dependence, model limitations, hallucination, automation bias and security concerns	Mandatory entering competency
Level C — Command supervision	Senior officers	Decide when AI should be used, challenged, overridden or disabled	Core war-college requirement
Level D — Technical specialization	Engineers, data scientists, acquisition and cyber specialists	Build, test, validate and maintain systems	Separate specialist education

The senior war colleges should concentrate principally on Level C because this is where command responsibility lies. They should not attempt to transform every senior officer into an AI developer, but neither should a commander be able to complete senior PME without demonstrating that he or she can supervise an AI-enabled staff.

This approach is reinforced by the June 2026 national-security AI policy framework, which explicitly preserves accountability through the chain of command even while accelerating access to advanced commercial and open-source models.

Human judgment should be evaluated explicitly, not assumed to survive AI adoption

The most significant pedagogical risk is not that officers will use AI, but that AI will become so productive that instructors cease to distinguish between the student’s judgment and the machine’s contribution.

The Army War College’s October 2026 strategic thought elective provides a useful emerging model because AI use is expected, documented and progressively integrated into assignments while students are explicitly required to challenge outputs and preserve human judgment. The instructor treats AI as an assistant rather than an authority and requires students to expose the process through which the machine contributed to their work. Strategic Education Powered by AI Integration — U.S. Army War College

This transparency principle should become institution-wide.

Every substantial AI-supported assignment should require four forms of disclosure:

Requirement	Purpose
AI system used	Establishes technological provenance
Prompting/workflow record	Reveals how the student structured the problem
Accepted and rejected outputs	Tests independent judgment
Final human rationale	Establishes ownership of the conclusion

The assessment should then focus on whether the student improves upon machine output rather than merely reproducing it.

This has become particularly urgent because Army War College faculty demonstrated in February 2026 that four leading commercial large-language models could pass a simulated War College comprehensive examination under existing criteria, even though their performance deteriorated under sustained questioning and researchers concluded that human judgment remained essential. Can AI Pass the U.S. Army War College Comprehensive Exam?

The implication is stark: any educational assessment that an AI system can satisfy without possessing responsibility, professional experience or genuine situational awareness is probably measuring only part of what senior military education intends to produce.

The 2026–2031 model should move from “AI permitted” to “AI audited”

Institutions have already passed the stage where simply allowing or banning generative AI is an adequate policy. The stronger model is audited use.

An audited-AI regime should distinguish three categories.

Category One — AI-required assignments

Students must use AI because learning how to supervise the system is the objective.

Category Two — AI-permitted assignments

Students can use AI, but must disclose its role and defend the final product.

Category Three — AI-excluded assessments

AI is deliberately unavailable because faculty need to measure independent recall, reasoning, oral judgment or unassisted synthesis.

This architecture prevents two opposite errors: treating AI as academic misconduct regardless of context, or allowing it to become an invisible co-author of every assessment.

Faculty capability is the largest institutional bottleneck

Curricular reform cannot proceed faster than faculty development. The Army War College’s March 2026 analysis of wargaming makes this clear by emphasizing that integrating games into PME is labor-intensive and pushes faculty beyond familiar teaching methods. Shall I Play a Game? Wargaming in PME

AI integration creates the same problem at larger scale. Faculty must understand enough about AI to design meaningful assignments, detect misuse, distinguish hallucinated from valid content, evaluate student-machine interaction and decide when technological complexity supports rather than distracts from learning objectives.

A 2026–2031 reform program should therefore measure faculty modernization directly.

Faculty modernization indicators

Indicator	2026–2031 target logic
Faculty with documented AI-literacy qualification	Should rise annually
Faculty trained in wargame design or facilitation	Should increase in departments using experiential learning
Courses using transparent AI-assessment rules	Should reach all programs where AI is permitted
Cross-disciplinary teaching teams	Should increase for technology-intensive subjects
Faculty exchanges with operational commands	Should increase current relevance
Faculty exchanges with industry/technical institutions	Should increase technical understanding
Research time protected for curriculum experimentation	Required for credible innovation
After-action reviews tied to curriculum revision	Should become routine

The principal danger is expecting faculty to modernize curriculum without changing workload, incentives or professional-development resources.

Experimental laboratories should become permanent institutional infrastructure

The next generation of senior PME institutions should have persistent experimentation environments rather than treating innovation as a temporary project. These environments need not all resemble physical technical laboratories. An educational laboratory can include secure AI sandboxes, wargaming suites, synthetic environments, classified planning cells, data-analysis workspaces and instrumented seminar rooms capable of capturing decision processes for later analysis.

The America’s AI Action Plan specifically calls for an AI and Autonomous Systems Virtual Proving Ground within the Department, beginning with the technical, geographic, security and resource requirements for such a facility. This concept has direct implications for PME because senior colleges are uniquely positioned to examine the interaction between technology and command rather than merely system performance.

The educational laboratory should differ from a weapons test facility. Its central question is not only “Does the system work?” but “How does the presence of the system change human decisions, command relationships, planning processes, delegation and strategic behavior?”

Proposed educational experimentation infrastructure

Facility/function	Purpose
Secure AI sandbox	Test multiple models without exposing sensitive data
Unclassified AI laboratory	Encourage broad experimentation and external collaboration
Classified planning environment	Connect education to current force problems
Wargaming suite	Test plans against adaptive opponents
Synthetic battlespace	Repeated experimentation at lower cost
Human-machine command laboratory	Study delegation and supervision
Decision instrumentation	Capture how officers reach decisions, not merely final answers
Red-team cell	Challenge technical and doctrinal assumptions
Data provenance environment	Train students to verify machine-supported conclusions

The key institutional design principle is interoperability among these facilities. A student should be able to develop a concept in seminar, test it in an unclassified game, refine it in a secure environment, examine technical implications with specialists and then defend the final judgment orally.

Metrics must shift from educational input to strategic performance

Current PME systems can easily count faculty, credits, courses, graduates and exercises. Those figures describe institutional activity but do not establish whether reform is producing better senior leaders.

CJCSI 1800.01G explicitly distinguishes direct and indirect assessment and defines institutional effectiveness as systematic evidence-based evaluation of how well an institution achieves its mission and learning outcomes. That framework should be used much more aggressively during 2026–2031.

A modernized measurement regime should contain five levels.

Level One — Inputs

Resources provided to education.

Level Two — Activities

What institutions actually do with those resources.

Level Three — Learning outputs

What students demonstrate at graduation.

Level Four — Transfer

Whether graduates apply those skills in subsequent assignments.

Level Five — Institutional effect

Whether the force receives better strategic advice, planning and command performance.

Proposed PME metric architecture

Metric level	Examples	Value	Limitation
Inputs	Faculty numbers, computing resources, wargaming facilities	Easy to measure	Says little about learning
Activities	AI-enabled courses, exercises, games, research projects	Tracks modernization	Activity can be mistaken for effectiveness
Learning outputs	Strategic reasoning scores, oral defense, adaptation performance	Measures immediate educational effect	Requires reliable assessment design
Transfer	Supervisor evaluations, staff performance, later command exercises	Tests real-world application	Attribution becomes harder
Institutional effects	Quality of joint plans, adaptive performance, strategic advice	Most important	Causal measurement extremely difficult

The system should therefore avoid claiming success merely because more AI courses, simulation hours or wargames have been introduced.

Reform indicators should be defined before the architecture is implemented

A serious 2026–2031 reform program needs observable indicators by which policymakers can determine whether integration is working.

Indicators of successful integration

Indicator	Positive interpretation
Strategy/history core remains stable while AI use expands	Integration rather than displacement
Students routinely challenge machine-generated conclusions	Human judgment retained
Wargames become embedded in assessment	Strategy tested through action
AI-supported assignments require provenance	Intellectual ownership remains visible
Classified and unclassified exercises are paired	Realism and contestability both preserved
Faculty development rises with technology adoption	Institutional capacity keeps pace
Graduates demonstrate stronger assumption management	Strategic reasoning improving
AI exercises include degraded, deceptive and contradictory data	Students prepared for adversarial conditions
Allied and interagency participation remains strong	Technology does not narrow institutional perspective
Multiple AI vendors/models are used	Reduces technological monoculture
After-action reviews lead to curriculum revision	Learning system becomes adaptive
Student work shows clearer distinction between machine output and professional judgment	AI literacy becoming command competence

Indicators of failed reform

Indicator	Warning meaning
Reduction in strategy/history content without measured gains	Curricular displacement
AI content confined to isolated electives	Superficial modernization
Students submit machine-generated work without process disclosure	Assessment validity degraded
Wargaming becomes technology demonstration	Pedagogical objective lost
Classified exercises dominate the curriculum	External challenge and intellectual diversity reduced
Faculty cannot explain AI tools used in their courses	Technology outruns teaching capacity
Single commercial platform becomes institutionally dominant	Vendor dependence
Rapid decision-making becomes a universal objective	Strategic deliberation undervalued
Technology metrics replace learning metrics	Reform becomes performative
Adversarial manipulation of AI is not exercised	Students trained only for benign system behavior

Curriculum should operate on a controlled replacement rule

One of the central governance problems is how to add new content without endlessly expanding programs. Every emerging topic generates legitimate claims on curricular time, but resident education cannot expand indefinitely.

A disciplined replacement rule should therefore apply:

No major new curricular requirement should be added unless faculty identify whether it is being integrated, substituted, moved to prerequisite education or removed elsewhere.

This simple requirement would force institutional transparency.

Four permissible pathways for new content

New requirement	Appropriate pathway	Example
Foundational for every strategic leader	Integrate into core	AI supervision
Important but rapidly changing	Deliver as recurring module	Current model capabilities
Highly specialized	Elective/specialist track	Machine-learning engineering
Outdated or redundant	Remove	Legacy technical detail no longer relevant

The most important category is integration. AI ethics, AI supervision, algorithmic bias, model deception and autonomous command should not all become separate courses. They should appear inside command, campaign planning, resource strategy, joint operations and wargaming.

Strategic foundations should have a minimum institutional floor

The protected-core principle should be translated into a governance mechanism. One option is a minimum percentage of resident senior PME devoted to durable strategic functions rather than current technical specialization.

This need not take the form of a rigid statutory quota, but institutions should report the distribution of instructional effort under common categories.

Proposed common reporting categories

Category	Definition
Strategy and policy	National strategy, military strategy, political objectives
Campaigning and joint operations	Theater strategy, operational design, joint integration
History and theory	Comparative strategic analysis
Leadership and command	Strategic leadership, mission command, ethics
Regional and adversary study	Geopolitical and cultural context
Resources and institutions	Industry, logistics, economics, acquisition
Technology and AI	Emerging technology, autonomy, cyber, data
Experiential learning	Wargames, simulations, exercises
Independent research	Thesis, research project, electives

A common annual reporting structure would make it possible to determine whether technology is being integrated or simply displacing older categories.

The system should deliberately preserve “slow thinking” environments

The technologically enabled military increasingly values decision speed, but an educational institution must also provide settings in which students can reason without operational tempo.

This is particularly important because current Army War College analysis warns that the drive to compress decision cycles through AI can risk converting iterative command processes into more linear automated action, increasing vulnerability to deception and escalation if commanders do not build restraint into the system before crisis. Untying the OODA Loop: Future Command and Decision Making

The curriculum should therefore include both:

• high-tempo simulation in which decisions must be made rapidly;
• slow analytical environments in which students are required to reconstruct assumptions, challenge evidence and examine second-order effects.

The purpose is not to privilege slow decisions over fast ones, but to teach officers when each is appropriate.

Reform should use paired exercises: acceleration followed by audit

One particularly strong educational method would be the paired exercise.

In Phase One, students operate in a high-tempo environment with AI-supported decision tools.

In Phase Two, the same decisions are audited after the exercise without time pressure.

The audit should ask:

• Which machine-generated recommendations were accepted?
• Which were rejected?
• What evidence changed the decision?
• Which assumptions proved wrong?
• Did speed improve or degrade outcome quality?
• Which decisions should have been delayed?
• Which decisions should have been delegated further?
• Where did the system create false confidence?
• What human information was unavailable to the model?
• Did the staff become dependent upon automated output?

This approach turns AI literacy into a measurable professional skill.

Multiple-model literacy should become part of senior education

The June 2026 national-security AI framework calls for onboarding advanced models from multiple vendors. The educational implication is significant: officers should not learn to equate AI with one interface or one commercial product.

Different systems can produce different answers, contain different guardrails, operate under different security assumptions and degrade differently under pressure. The Army War College’s 2026 experiment comparing four major commercial models demonstrated precisely that point because all passed the assessment but performed differently and showed degradation during extended interaction.

A senior officer should therefore be trained to ask:

Question	Why it matters
Do multiple models agree?	Agreement can increase confidence but does not establish truth
Where do models diverge?	Divergence can reveal hidden assumptions
What data environment supports each?	Determines suitability
What model is authorized for classified use?	Security and policy requirement
What functions are deterministic versus generative?	Affects reliability
Can the output be reproduced?	Matters for staff accountability
Can the system explain evidence provenance?	Essential for command trust
How does the model behave under adversarial prompting?	Relevant to operational security

This is the AI equivalent of understanding different intelligence sources rather than relying on one channel.

The architecture should protect coalition education from technological fragmentation

A heavily classified, technologically specialized PME environment can create an unintended interoperability problem: U.S. officers may gain access to sophisticated systems that many allies cannot use, while multinational military effectiveness continues to depend upon common understanding and shared planning.

The Army War College’s Joint Warfighting Program explicitly identifies international officers as a competitive advantage and uses an unclassified operational wargaming system in part because such an environment allows meaningful multinational participation.

The Naval War College similarly uses wargaming for multinational professional development; its current wargaming program supports broad educational and analytical activity rather than purely U.S.-only classified experimentation.

The 2026–2031 architecture should therefore preserve a substantial coalition-accessible educational layer, even as classified U.S. experimentation expands.

Coalition-access architecture

Educational level	Participation
Strategic theory and policy	Broad multinational
Unclassified operational games	Broad multinational
Controlled technical demonstrations	Selected partners
Classified planning	Access according to national disclosure rules
Sensitive AI systems	Restricted
Coalition interoperability exercises	Designed around lowest necessary classification level

The objective should be to prevent technological sophistication from producing strategic isolation.

Industry engagement should expand, but vendor capture should be actively controlled

Because major AI development increasingly occurs in the commercial sector, senior military institutions need structured interaction with industry. The White House’s national-security AI policy explicitly calls for rapid onboarding of advanced commercial and open-source technology, while the broader AI Action Plan emphasizes close cooperation between government and industry.

The educational value is clear: officers need direct exposure to technical possibilities and constraints rather than filtered descriptions.

However, war colleges should not become product demonstration environments.

The proper model is competitive exposure:

• multiple firms;
• open-source alternatives;
• government-developed systems;
• structured comparison;
• faculty-defined learning objectives;
• disclosure of commercial relationships;
• no single-vendor dependency for core assessments.

Policy pathways for 2026–2031

Pathway A — Incremental integration

Institutions retain the present curriculum and progressively embed AI, autonomy and digital tools into existing courses.

Authority: existing service and Joint Staff educational governance.

Expected effect: low-disruption modernization.

Implementation burden: moderate.

Time to effect: short.

Reversibility: high.

Principal downside: uneven implementation and risk of superficial integration.

Pathway B — Protected-core restructuring

The Department defines mandatory strategic competencies and allows institutions to redesign all other content around them.

Authority: CJCS PME policy and service education governance, potentially reinforced through statutory oversight.

Expected effect: stronger protection against curricular displacement.

Implementation burden: high during transition.

Time to effect: medium.

Reversibility: medium.

Principal downside: poorly designed protection could become bureaucratic rigidity.

Pathway C — Laboratory-centered PME

Senior colleges receive permanent secure experimentation infrastructure, AI sandboxes, wargaming capacity and decision-analysis environments.

Authority: Department and service resourcing, aligned with the AI Action Plan’s virtual proving-ground concept.

Expected effect: significantly more experiential learning and technology testing.

Implementation burden: high.

Time to effect: medium to long.

Reversibility: low after infrastructure investment.

Principal downside: technology infrastructure can dominate educational purpose.

Pathway D — Performance-based PME

Traditional course completion remains necessary but strategic competence is certified through integrated written, oral, wargaming and AI-supervision assessments.

Authority: Joint Staff quality-management and accreditation mechanisms.

Expected effect: shifts incentives from course delivery toward demonstrated capability.

Implementation burden: high because valid assessment instruments must be designed.

Time to effect: medium.

Reversibility: medium.

Principal downside: complex competencies can be distorted if reduced to simplistic scoring.

Pathway E — Distributed lifelong strategic education

Resident PME becomes one stage in a continuous system incorporating prerequisite AI literacy, post-graduation refreshers, digital simulations and recurring strategic exercises.

Authority: service education and talent-management systems.

Expected effect: reduces pressure on resident curriculum and addresses rapid technological depreciation.

Implementation burden: high institutional coordination.

Time to effect: long.

Reversibility: medium.

Principal downside: fragmented participation and inconsistent follow-through.

A combined architecture is more plausible than a single pathway

The public record supports a hybrid rather than a singular transformation. The Joint Staff already provides quality-management architecture; service colleges are experimenting with AI and wargaming; Air War College retains an explicitly broad strategic curriculum; Naval War College is producing classified AI experimentation; Army War College is integrating AI into both wargaming and academic assignments; and national policy is pushing Senior Military Colleges toward deeper AI literacy and research.

The most defensible 2026–2031 institutional model therefore combines:

protected strategic foundations + embedded AI literacy + classified and unclassified experimentation + permanent wargaming + multi-model exposure + faculty modernization + performance-based assessment + recurring post-resident education.

Principal risk register

Risk	Mechanism	Early indicator	Institutional response
Curricular displacement	New technology hours remove strategic content	Declining strategy/history allocation	Protected competency floor
AI dependency	Students cannot perform without machine support	Weak unaided oral defense	AI-excluded assessments
Automation bias	Machine recommendations accepted too easily	Limited rejection of AI outputs	Required challenge protocols
Faculty lag	Students understand tools better than instructors	Inconsistent AI policy	Faculty development
Classified intellectual closure	Assumptions become insulated from challenge	Growth in classified hours without open counterpart	Paired unclassified analysis
Vendor capture	One provider dominates education	Single-model dependency	Multi-vendor policy
Wargame theater	Exercises prioritize spectacle over learning	Poor linkage to outcomes	Assessment-driven game design
Metric gaming	Institutions optimize visible indicators	Rising activity without learning evidence	Multi-level evaluation
Technological obsolescence	Curriculum tied to current platforms	Content aging within one academic cycle	Modular refresh system
Coalition fragmentation	Allies excluded from advanced education	Reduced multinational participation	Coalition-access layer
Loss of historical depth	Current conflict becomes dominant template	Narrow case selection	Comparative case requirements
Excessive decision-speed culture	Rapid action treated as universal virtue	Little deliberate decision practice	Variable-tempo exercises

Institutional metrics for 2026–2031

A credible reform program should publish a small, stable set of comparable metrics rather than hundreds of internal indicators.

Recommended institutional scorecard

Metric	Measurement	Desired direction
Strategic-core share	Percentage of curriculum tied to protected competencies	Stable
AI-integrated core courses	Share of core courses using AI analytically	Rising
Wargaming participation	Student hours in adversarial decision exercises	Rising
AI provenance compliance	Percentage of AI-supported work with full disclosure	Near-universal
Faculty AI qualification	Faculty meeting defined literacy standard	Rising
Multi-model exposure	Number of distinct authorized model families used	More than one
Classified/unclassified balance	Ratio of experimental learning environments	Balanced according to mission
Adversarial AI exercises	Number of exercises involving deception or corrupted data	Rising
Human-judgment assessment	Share of graduation assessment involving oral/adaptive performance	Rising
Curriculum refresh cycle	Time required to update technology modules	Falling
Alumni transfer assessment	Evidence of application in later assignments	Increasing coverage
Coalition participation	International/interagency participation in accessible exercises	Stable or rising

No single metric should become a score of institutional quality. The value lies in trends and trade-offs.

Decision thresholds

The architecture should trigger formal review if any of the following conditions appear.

Threshold One: strategic/history/policy content falls materially for two consecutive academic cycles without direct evidence that equivalent competencies are being achieved elsewhere.

Threshold Two: more than half of major assessments permit unrestricted AI use without corresponding oral or process-based validation.

Threshold Three: a single commercial AI provider becomes functionally indispensable to a senior college.

Threshold Four: classified learning expands significantly while coalition and unclassified experimentation decline.

Threshold Five: faculty AI-literacy rates fail to keep pace with mandatory student AI use.

Threshold Six: wargaming hours expand but are not mapped to formal learning outcomes.

Threshold Seven: graduate assessment measures technical familiarity more often than strategic judgment.

These thresholds would not automatically establish failure, but they would justify institutional review.

The architecture should be designed around replacement, redundancy and resilience

Military institutions often discuss resilience in terms of networks and logistics, but professional education also requires resilience.

The first form is cognitive redundancy. Officers should be capable of performing core reasoning tasks when AI assistance becomes unavailable.

The second is model redundancy. Institutions should avoid dependence upon one AI architecture.

The third is data redundancy. Students should learn to compare machine output with independent sources.

The fourth is methodological redundancy. Strategic conclusions should survive more than one analytical method.

The fifth is institutional redundancy. Classified, unclassified, military, interagency and multinational perspectives should all remain present in the educational ecosystem.

An architecture that maximizes technological sophistication but removes these redundancies would be efficient under normal conditions and brittle under disruption.

What successful reform would look like by 2031

By the end of the proposed horizon, success should not be defined by whether every war college has created an AI department or dramatically increased the number of technology courses.

Successful reform would be visible in a more subtle institutional pattern.

Senior officers would enter resident education with baseline AI literacy rather than spend scarce time learning elementary terminology.

AI would be used routinely but transparently across strategy, campaigning, leadership, resource analysis and research.

Students would be expected to challenge AI outputs as a normal professional responsibility.

Wargaming would be a core method of demonstrating judgment rather than a peripheral exercise.

Classified operational experimentation would be balanced by unclassified comparative reasoning.

Faculty would possess enough technical literacy to supervise AI-enabled learning rather than outsource it to specialists.

Strategic history and theory would remain central because technology would be used to interrogate them rather than replace them.

Graduation would require officers to demonstrate that they can reason both with and without machine assistance.

Curricula would update technological modules rapidly while preserving durable strategic outcomes.

Most importantly, the system would judge reform through evidence of better command reasoning under uncertainty, rather than through the visibility of new technology.

Key judgments

The first judgment is that the strongest 2026–2031 educational architecture is integration rather than curricular substitution. The external policy environment requires faster AI adoption and stronger warfighting relevance, but current Joint Staff policy and service-college practice provide no institutional necessity to sacrifice strategic foundations in order to satisfy that requirement.

The second judgment is that strategic foundations should be protected through competency floors rather than immutable course structures. This allows institutions to modernize content aggressively without allowing core capabilities—strategic framing, campaigning, joint integration, command judgment, history and resource reasoning—to disappear under technological pressure.

The third judgment is that wargaming should become part of the formal assessment system because written analytical competence alone is insufficient to reveal whether an officer can adapt when an adversary changes the problem. Army and Naval War College evidence shows that gaming is already sufficiently mature to perform this role if faculty capacity and assessment design are strengthened.

The fourth judgment is that AI literacy for senior leaders should culminate in command supervision, not technical engineering. The central professional requirement is knowing when to employ, question, override, constrain or reject machine output while retaining accountability.

The fifth judgment is that classified experimentation should expand, but never become the sole intellectual environment for modern PME. The system requires an unclassified layer for comparative scholarship, coalition education and contestability.

The sixth judgment is that faculty capability is the pacing constraint. A college cannot integrate AI, gaming and modern simulation at a higher level than its faculty can meaningfully supervise.

The seventh judgment is that assessment validity is now a strategic issue. The ability of advanced AI systems to satisfy traditional academic rubrics means that professional military institutions must increasingly test intellectual ownership, adaptation, judgment and responsibility rather than polished written output alone.

The eighth judgment is that the future war college should function simultaneously as school, laboratory and assessment environment, but its educational mission must remain dominant over technological experimentation.

What would change the assessment

The present judgment would strengthen if Joint Staff accreditation and quality-management processes adopt common metrics for AI integration, wargaming, human judgment and curricular balance; if service colleges publish comparable data on technology integration without showing corresponding erosion of strategy and history; if longitudinal evidence links experiential and AI-enabled education to stronger graduate performance; and if classified experimentation continues to coexist with substantial multinational and unclassified learning.

The assessment would weaken if technology courses materially displace strategic education across multiple institutions, if commercial AI systems become opaque dependencies inside core assessment, if AI-generated products cannot be distinguished from student reasoning, or if increased classification significantly narrows the diversity of the educational ecosystem.

A major reassessment would also be required if validated empirical evidence showed that specific strategic competencies currently associated with resident PME can be developed more effectively through distributed, simulation-based or AI-mediated education at substantially lower cost.

Open official record

The public record does not yet provide a Department-wide dataset showing the proportion of senior PME curricula devoted to AI, history, strategic studies, campaigning, wargaming, classified exercises and technology instruction under standardized definitions. Without such data, claims of either widespread curricular displacement or complete preservation remain difficult to establish.

There is also no public longitudinal dataset establishing whether graduates of AI-integrated senior PME subsequently perform better in combatant commands, joint staffs or strategic advisory positions than graduates educated through previous models.

The public record does not establish a common Department-wide definition of what level of AI literacy senior officers must demonstrate at graduation, despite strong policy direction toward expanded AI education.

The extent of classified AI experimentation across senior colleges is necessarily only partially observable, preventing a complete comparison of institutional maturity.

Finally, the evidence base remains insufficient to determine whether AI-enabled educational tools will improve long-term strategic reasoning or merely accelerate near-term production. Current Army and Naval War College experiments demonstrate institutional feasibility and important learning opportunities, but they do not yet establish long-term causal effects on command performance.

---

Copyright of debuglies.com – Even partial reproduction of the contents is not permitted without prior authorization Reproduction reserved