Abstract: The accelerating pace of digitalisation and autonomous systems challenges traditional officer education. Future leaders must be prepared to operate in environments shaped by artificial intelligence, automation, and complex ethical dilemmas. This paper explores how AI‑driven simulation can enhance readiness and adaptability in officer training, ensuring that cadets develop both technical competence and strategic judgment. Current training models often fail to replicate the unpredictability of modern battlefields. Without integrating emerging technologies, officer education risks lagging behind adversaries who adopt them aggressively. The absence of adaptive, tech-driven methods limits cadets’ opportunities to practice decision-making under uncertainty and to confront the ethical and legal questions raised by autonomous systems.
Problem statement: How can military education move beyond static, theory‑heavy instruction to integrate AI‑driven simulations that realistically prepare officers for complex, tech‑driven operations?
So what?: Military academies and defence institutions must actively integrate AI‑driven simulations into officer curricula, while researchers and technologists collaborate to design adaptive frameworks that connect doctrinal theory with realistic, practice‑oriented training. This joint effort ensures future officers are prepared for complex, tech‑driven operations.
An Accelerating Pace of Digitalisation
The accelerating digitalisation of defence and the rapid integration of artificial intelligence (AI) and autonomous systems are reshaping the character of contemporary military operations. Modern battlefields now feature compressed decision cycles, data‑intensive command‑and‑control processes, and continuous interaction between human operators and intelligent systems. These conditions impose new cognitive, ethical, and professional demands on junior officers, who must exercise judgment and accountability in environments shaped by automation, algorithmic decision support, and operational uncertainty. As a result, military education systems must adapt their training models to prepare officers for leadership in technologically mediated contexts.[1]
Traditional officer education—centred on theory‑heavy instruction, doctrinal study, and scripted exercises—remains essential for establishing foundational knowledge and military ethos. However, these methods struggle to replicate the complexity, unpredictability, and ethical ambiguity of modern operations. Classroom‑based learning provides limited opportunities for cadets to experience the consequences of decision‑making under uncertainty or to engage meaningfully with emerging technologies. This gap between doctrinal understanding and practical readiness risks undermining the ability of newly commissioned officers to operate effectively in digitalised environments.[2]
The challenge is most acute for AI‑enabled and autonomous systems. Officers are increasingly expected to supervise intelligent technologies while retaining human authority, accountability, and ethical responsibility. Without training environments that mirror these realities, officer education risks falling behind operational developments and adversary capabilities. While simulation‑based training has long been used to bridge theory and practice, conventional systems rely on static scenarios and predetermined outcomes that fail to capture the adaptive, uncertain, and adversarial nature of contemporary conflict.[3] To address this gap, this paper proposes the use of AI‑enabled, dynamically adaptive simulation environments capable of generating evolving scenarios, modelling adversary behaviour, and exposing cadets to realistic decision‑making pressures.
Background
Officer education has continually evolved in response to shifts in warfare, technological development, and societal expectations of military leadership. Historically, military academies emphasised doctrinal instruction, discipline, and the transmission of professional norms as the foundation of officer formation. These approaches were not based on predictable or stable battlefields, even ancient and early‑modern commanders operated under profound uncertainty—but rather on the organisational and technological realities of their time, including slower information flows, limited situational awareness, and hierarchical command structures. Over time, experiential elements such as field exercises and war gaming were introduced to complement theoretical instruction and develop practical decision-making skills.[4]
Simulation has played a central role in this pedagogical evolution. Early forms of military simulation, such as map exercises and manual war games, enabled officers to rehearse tactical decisions and explore operational concepts without the risks of live manoeuvres. With advances in computing, digital simulations became increasingly capable of modelling complex systems, large‑scale operations, and multi‑domain interactions. Contemporary military training now employs a wide spectrum of simulation tools—from immersive virtual reality environments used in pilot and infantry training to constructive simulations supporting staff planning and joint operations. These systems have demonstrated value in improving learning efficiency, reducing training costs, and enabling safe experimentation.[5], [6]
Despite these advances, many existing simulation systems remain constrained by scripted scenarios and limited adaptability. Trainees often encounter predefined decision paths and predictable outcomes, which can reduce realism and limit opportunities for genuine problem‑solving. As a result, simulation‑based training may reinforce procedural compliance rather than fostering adaptive judgment, initiative, and responsibility. This limitation has prompted growing interest in applying artificial intelligence and machine learning to training systems.[7]
AI‑enabled training environments offer the potential to generate dynamic scenarios, model intelligent adversaries, and adapt training content in response to trainee behaviour. Machine‑learning techniques can analyse performance data to provide personalised feedback and adjust scenario complexity in real time. Educational research suggests that such adaptive environments support deeper learning by engaging trainees in iterative cycles of action, feedback, and reflection.[8], [9]
However, the integration of AI into military training also introduces important limitations and risks. These include challenges related to transparency and explainability, the potential for automation bias, and concerns about over-reliance on algorithmic recommendations. Ethical, legal, and doctrinal considerations remain central: military organisations emphasise the primacy of human judgment and accountability, particularly in decisions involving the use of force or the supervision of autonomous systems. Training environments must therefore reinforce ethical reasoning and legal compliance rather than delegating authority to algorithms.[10] Scholars caution that uncritical adoption of AI technologies may erode professional responsibility if human oversight is diminished. Consequently, AI‑driven simulation must be designed as a tool to support—rather than replace—human leadership.[11], [12]
Conceptual Foundations
The integration of AI into military training requires a clear conceptual foundation that links technological capabilities with concrete educational needs. In this context, AI-driven simulation refers to training environments that use artificial intelligence and machine-learning techniques to generate adaptive scenarios, model complex operational systems, and respond dynamically to trainees’ decisions. Unlike traditional scripted simulations, which follow predetermined branches, AI‑enabled systems can modify environmental variables, adversary behaviour, and scenario trajectories in real time, producing genuinely emergent learning experiences.[13]
A practical illustration highlights the difference. In many current tactical simulations, a cadet confronted with an ambush may be offered only three predefined options: advance, withdraw, or call for fire support. Regardless of the trainee’s reasoning, the scenario progresses along one of these fixed paths. An AI‑enabled system, by contrast, could generate an adversary response based on the cadet’s timing, formation, communication discipline, or sensor use. It could escalate, disperse, feint, or exploit a detected weakness, creating uncertainty that mirrors real operational dynamics. This example demonstrates how AI can transform simulations from linear rehearsals into adaptive environments that cultivate judgment under pressure.
Adaptive learning is closely linked to these capabilities. In officer education, adaptive systems can analyse performance patterns, identify cognitive or procedural weaknesses, and adjust scenario difficulty accordingly. This supports differentiated instruction while maintaining common professional standards. When combined with instructor oversight and structured debriefing, adaptive learning enhances engagement and promotes reflective practice.[14]
Decision‑making under uncertainty remains a core competency for modern officers, who must operate with incomplete information, time pressure, and technologically mediated situational awareness. Uncertainty arises not only from adversary behaviour but also from the interaction between human judgment and automated systems. Officers must interpret probabilistic information, recognise the limits of algorithmic tools, and remain accountable for decisions even when supported by AI‑generated recommendations. Training environments that expose cadets to such uncertainty reinforce ethical reasoning, responsibility, and resilience—competencies essential for leadership in human–machine teams.[15]
The pedagogical foundations of simulation‑based learning further support the integration of AI. Experiential learning theory emphasises cycles of action, reflection, and conceptualisation; deliberate practice highlights repeated, goal‑oriented performance with immediate feedback; and cognitive load theory underscores the need to balance realism with instructional clarity.[16], [17], [18] AI‑enabled systems can operationalise these principles by providing tailored feedback, adjusting scenario complexity, and ensuring that technological sophistication enhances rather than overwhelms learning.
Finally, the operational environment continues to shape the competencies required of junior officers. Modern operations demand digital literacy, ethical awareness, and the ability to lead within human–machine teams. Officers must interpret data‑driven recommendations without succumbing to automation bias, maintain accountability for decisions, and operate effectively across joint and multi‑domain contexts. AI‑driven simulation offers a controlled yet realistic environment in which these competencies can be developed before officers enter operational service.[19]
Problem Analysis
A key limitation of current training models is that they often present cadets with decision problems that have only a small number of predetermined “correct” options. For example, in many tactical simulations, an ambush scenario offers three fixed choices—advance, withdraw, or request fire support—regardless of the cadet’s timing, formation, communication discipline, or use of sensors. This structure teaches procedural correctness rather than adaptive judgment.[20], [21] An AI‑enabled system, by contrast, could generate adversary responses based on the cadet’s behaviour: an enemy force might disperse, escalate, feint, or exploit a detected weakness. This shift from fixed branches to emergent behaviour exposes cadets to realistic uncertainty and forces them to reason through the consequences of their actions.[22] It also clarifies the nature of ethical dilemmas in technologically mediated operations. The issue is not that AI has “lower ethical standards,” but that officers must make accountable decisions in situations where algorithmic tools influence perception, timing, and risk assessment.[23] The real challenge—and the reason this matters—is ensuring that future officers can recognise when to rely on automated support, when to question it, and how to maintain responsibility for decisions in environments shaped by intelligent systems.[24]
Proposed Framework for AI‑Driven Simulation Integration
While AI‑enabled simulation has made significant progress, it is essential to distinguish between capabilities that already exist and those that remain developmental. Several military organisations have implemented AI‑assisted training systems that demonstrate the feasibility of adaptive, data‑driven simulation. The United States Air Force’s Pilot Training Next (PTN) programme uses machine‑learning analytics to assess trainee performance and adjust scenario difficulty in real time, improving learning efficiency and situational awareness while highlighting the continued need for instructor oversight.[25] DARPA’s AlphaDogfight Trials have shown how AI‑controlled adversaries can generate dynamic and unpredictable behaviour for pilot training, though these systems remain limited by data availability, scenario scope, and the challenge of validating AI behaviour against doctrinal standards.[26] Similarly, the UK Ministry of Defence’s Synthetic Training Environment integrates AI to create evolving battlefield conditions, demonstrating operational potential while also revealing constraints in interoperability and scenario fidelity.[27] These examples illustrate that the capabilities described in this framework are not purely speculative; however, more advanced forms of fully emergent, continuously learning simulations remain hypothetical but technically plausible, representing the next stage of development as the technology matures.[28]
This distinction also clarifies the ethical dimension. The issue is not that AI possesses “lower ethical standards,” but that human judgment becomes more difficult when algorithmic tools influence perception, timing, and risk assessment.[29] Officers must remain accountable even when supported by automated systems, and training must therefore cultivate the ability to recognise when to rely on AI, when to question it, and how to maintain responsibility for decisions in technologically mediated environments. AI‑enabled simulation provides a controlled setting in which cadets can practise navigating uncertainty, automation bias, and ethical dilemmas before encountering them in real operations.[30]
Use Cases and Illustrative Scenarios
The practical value of AI‑driven simulation in officer education becomes most evident when examined through concrete training applications. Existing military programmes already demonstrate elements of adaptive simulation, while emerging technologies point toward future possibilities. Use cases and illustrative scenarios show how AI‑enabled environments can translate doctrinal instruction into experiential learning, enabling cadets to practise decision‑making, leadership, and ethical judgment in realistic contexts. These scenarios are designed not as isolated exercises but as integrated components of a broader educational framework, reinforcing learning objectives across multiple domains.[31]
A prominent application of AI‑assisted simulation is decision‑making under uncertainty at the tactical level. The United States Air Force’s Pilot Training Next (PTN) programme provides a relevant example: machine‑learning analytics assess trainee performance and adjust scenario difficulty in real time, requiring pilots to respond to evolving conditions rather than predetermined outcomes.[32] In officer education, similar scenarios place cadets in command roles where they must interpret incomplete information, allocate limited resources, and respond to adversary actions that adapt dynamically to their decisions. AI‑enabled adversary models—demonstrated in DARPA’s AlphaDogfight Trials—adjust tactics based on trainee behaviour, introducing unpredictability and forcing cadets to reassess assumptions.[33] This approach encourages critical thinking and reinforces the importance of situational awareness, risk assessment, and accountability. Instructor‑led debriefings connect these experiences to doctrinal principles, ensuring that practical lessons remain grounded in theoretical understanding.
Human–machine teaming scenarios represent another critical use case. As modern operations increasingly involve collaboration between human operators and intelligent systems, officers must learn to interpret algorithmic recommendations while retaining decision authority. The UK Ministry of Defence’s Synthetic Training Environment (STE) already incorporates AI‑supported decision aids that provide probabilistic assessments and sensor‑fusion outputs.[34] AI‑driven simulations in officer education can model similar decision‑support systems, requiring cadets to critically evaluate automated inputs, balance technological assistance with human judgment, and recognise the risks of automation bias. Such scenarios reinforce the necessity of maintaining human oversight in AI‑supported operations.
Ethical dilemmas involving autonomous systems are particularly well-suited to simulation‑based training. AI-enabled environments can present cadets with scenarios involving the use of force, the presence of civilians, and autonomous platform behaviour under ambiguous conditions. NATO’s Allied Command Transformation (ACT) has experimented with AI‑enabled wargaming that embeds ethical and legal considerations into operational decision‑making.[35] By situating ethical challenges within realistic contexts, simulations move beyond abstract discussion and require cadets to confront the consequences of their decisions. Instructor facilitation ensures that ethical reasoning is aligned with international humanitarian law, rules of engagement, and institutional values, fostering moral responsibility and professional integrity.
Tactical and operational vignettes further illustrate the versatility of AI‑driven simulation. At the tactical level, cadets may engage in platoon‑ or company‑level operations involving coordination, communication, and leadership under stress. At the operational level, simulations can support staff training by modelling joint and multi‑domain environments, enabling cadets to explore the interplay between land, air, maritime, cyber, and information domains. Programmes such as the Australian Defence Force’s Joint Collective Training initiatives demonstrate how AI‑supported constructive simulations can replicate complex, multi‑domain interactions. These vignettes reinforce the interconnected nature of modern operations and prepare officers for the complexity of contemporary command environments.
Across these use cases, AI‑driven simulation supports iterative learning through repetition and variation. Cadets can revisit scenarios with modified parameters, allowing them to test alternative approaches and observe different outcomes. This iterative process aligns with deliberate‑practice principles and supports the development of expertise over time. While some advanced capabilities remain developmental, existing programmes across allied militaries demonstrate that adaptive, AI‑assisted simulation is already operationally viable and continues to evolve.[36]
Implementation Considerations
The successful integration of AI‑driven simulation into basic officer education depends not only on technological capability but also on careful consideration of institutional, pedagogical, and organisational factors. Implementation must be approached as a systemic transformation, because AI‑enabled simulation affects curriculum design, instructor roles, assessment practices, data governance, and organisational culture—not merely the technical infrastructure. Aligning these elements is essential to ensure that AI‑enabled training enhances, rather than disrupts, established educational processes and professional norms.
From a technical perspective, AI‑driven simulation environments require robust computational infrastructure capable of supporting real‑time data processing, adaptive scenario generation, and performance analytics. These systems must be reliable, scalable, and interoperable with existing simulation platforms and learning management systems used by military academies. Interoperability is particularly important for ensuring continuity between classroom instruction, simulation‑based training, and field exercises. Examples include integrating AI‑enabled components into existing constructive simulation suites or linking adaptive scenarios to established virtual training platforms. Modular system design allows institutions to introduce AI capabilities incrementally, reducing technical risk and facilitating adaptation to evolving requirements.
Instructor roles and human oversight remain central to the effective use of AI‑driven simulation. Although adaptive systems can automate scenario generation and feedback delivery, instructors are essential for contextualising learning, guiding reflection, and reinforcing professional values. Educators must therefore be trained not only in the technical operation of simulation systems but also in interpreting performance data, identifying patterns in trainee behaviour, and facilitating meaningful debriefings. This human‑centred approach ensures that AI‑enabled simulation supports—rather than diminishes—the development of responsibility, autonomy, and ethical judgment in officer cadets.[37]
Data governance, security, and privacy are critical considerations in implementing AI‑enabled training systems. Simulation environments generate extensive performance data, including decision patterns, communication behaviours, and cognitive indicators, which must be managed in accordance with institutional policies and legal frameworks. Transparent data‑governance practices are essential for maintaining trust among cadets and instructors, particularly when performance data is used for assessment or progression decisions. Cybersecurity measures must also be implemented to protect sensitive training data and prevent unauthorised access, manipulation, or exploitation—especially given the increasing integration of networked simulation systems across military institutions.[38]
Interoperability with existing military education systems extends beyond technical compatibility to include doctrinal and curricular alignment. AI‑driven simulations must reflect current doctrine, rules of engagement, and operational concepts to ensure relevance and credibility. Regular updates and validation processes are necessary to maintain alignment with evolving operational requirements. Collaboration among educators, technologists, and subject-matter experts supports the continuous refinement of simulation content and learning objectives, ensuring that AI-enabled training remains pedagogically sound and operationally realistic.[39]
Challenges in adoption are not limited to technical factors. Cultural and institutional resistance may arise from concerns about over‑reliance on technology, perceived threats to traditional teaching roles, or uncertainty about the reliability and transparency of AI‑enabled systems.[40] Addressing these concerns requires clear communication of the purpose and limitations of AI, professional development for instructors, and demonstration of the added value that adaptive simulation brings to officer education. When supported by leadership commitment and organisational learning, these measures help foster a culture that views AI‑enabled simulation as a complement to—not a replacement for—human expertise.
Evaluation and Assessment
The effectiveness of AI‑driven simulation in officer education depends on rigorous evaluation and assessment mechanisms that align training outcomes with educational objectives and operational requirements. Evaluation must extend beyond technical system performance to encompass learning effectiveness, decision quality, ethical judgment, and the development of professional competencies. A comprehensive assessment approach ensures that AI‑enabled training contributes meaningfully to officer formation rather than serving as a technological novelty.
Measuring training effectiveness in AI‑driven simulation environments requires a combination of qualitative and quantitative methods. Performance data generated during simulation exercises provides objective indicators of decision‑making patterns, response times, and outcome effectiveness. These metrics are typically analysed by instructors, assessment boards, and curriculum committees, who interpret the data in relation to doctrinal standards and learning objectives. Structured debriefings, instructor observations, and reflective self‑assessments complement quantitative data by capturing cognitive, ethical, and emotional dimensions of learning. Such mixed‑method approaches support a holistic understanding of cadet development and reinforce the central role of instructors in interpreting performance data.[41]
Metrics for readiness, adaptability, and decision quality are central to evaluating AI‑driven simulation outcomes. Readiness can be assessed through scenario‑based evaluations that measure a cadet’s ability to apply doctrinal knowledge under realistic conditions. Adaptability is reflected in the capacity to adjust strategies in response to changing scenarios and adversary behaviour. Decision quality encompasses not only operational effectiveness but also adherence to ethical and legal standards. By mapping these metrics to defined officer competencies, educators and assessment boards can evaluate progress systematically and identify areas requiring further development.[42]
Validation of AI‑driven simulations is essential to ensure credibility and educational value. Validation processes—conducted by subject‑matter experts, doctrine centres, simulation specialists, and external evaluation bodies—confirm that simulated scenarios accurately reflect doctrinal principles, operational realities, and ethical constraints. This includes verifying the behaviour of AI‑generated adversaries, the accuracy of environmental modelling, and the reliability of performance analytics. Continuous validation ensures that simulations remain aligned with evolving operational requirements and technological developments. Safeguards must also be implemented to monitor for automation bias, ensuring that cadets maintain critical judgment and do not over-reliance on algorithmic recommendations.[43]
Finally, assessment mechanisms must support institutional learning. Aggregated performance data can inform curriculum development, instructor training, and resource allocation. By identifying systemic patterns—such as recurring weaknesses in ethical reasoning, situational awareness, or decision‑making under uncertainty—military academies can refine their educational strategies and strengthen officer preparation.[44] Institutional learning processes, supported by regular review cycles and cross‑departmental collaboration, ensure that AI‑enabled simulation contributes not only to individual cadet development but also to the continuous improvement of the officer education system as a whole.
Conclusion
The integration of AI‑driven simulation into basic officer education represents a significant opportunity to strengthen the preparation of future military leaders. The analysis presented in this paper demonstrates that AI‑enabled training is not merely a speculative concept but an emerging reality, with programmes such as the United States Air Force’s Pilot Training Next, DARPA’s AlphaDogfight Trials, and the UK Ministry of Defence’s Synthetic Training Environment already illustrating the potential of adaptive, data‑driven simulation. These initiatives show that AI can enhance decision‑making under uncertainty, support human–machine teaming, and embed ethical dilemmas within realistic operational contexts.
At the same time, the full potential of AI‑driven simulation remains developmental. More advanced forms of continuously learning, fully emergent simulation environments are technically plausible but not yet operational. This distinction underscores the need for a balanced approach that integrates existing capabilities while preparing for future advancements. Effective implementation requires systemic transformation across technical infrastructure, instructor training, data governance, and curricular alignment. AI‑enabled simulation must reinforce—not replace—human judgment, ethical responsibility, and professional accountability.
The proposed framework offers a structured pathway to integrate AI‑driven simulation into officer education through a modular architecture, scenario variability, adaptive feedback loops, and embedded ethical and legal decision‑making modules. These elements support experiential learning, deliberate practice, and the development of core officer competencies. Rigorous evaluation and validation processes, conducted by instructors, assessment boards, doctrine centres, and simulation specialists, ensure that AI‑enabled training remains credible, doctrinally aligned, and pedagogically sound.
Ultimately, AI‑driven simulation should be understood as a tool that enhances the capacity of military education systems to prepare officers for the complexity of contemporary operations. By providing controlled yet realistic environments in which cadets can practise decision‑making, confront ethical dilemmas, and develop resilience under uncertainty, AI‑enabled simulation contributes to the formation of adaptable, responsible, and technologically literate leaders. As military organisations continue to modernise, the integration of AI‑driven simulation will play an increasingly important role in ensuring that officer education remains aligned with operational realities and future challenges.
[1] NATO, NATO Artificial Intelligence Strategy (Brussels: NATO Headquarters, 2021).
[2] Leonard Wong and Stephen Gerras, Lying to Ourselves: Dishonesty in the Army Profession (Carlisle, PA: Strategic Studies Institute, 2015).
[3] Paul Scharre, Army of None: Autonomous Weapons and the Future of War (New York: W. W. Norton, 2018).
[4] Leonard J. Matthews, “The Future of Military Education,” Parameters 49, no. 2 (2019): 5–16.
[5] Matthew B. Caffrey, On Wargaming: How Wargames Have Shaped History and How They May Shape the Future (Newport, RI: Naval War College, 2000).
[6] J. D. Fletcher and A. P. Wind, Costs, Benefits, and Effectiveness of Military Training (Santa Monica, CA: RAND Corporation, 2013).
[7] Eduardo Salas et al., “The Science of Training and Development in Organizations,” Psychological Science in the Public Interest 13, no. 2 (2012): 74–101.
[8] Sebastian Riedel et al., “Adaptive Simulation Environments for Military Training,” Simulation & Gaming 51, no. 4 (2020): 475–495.
[9] Chris Dede, “The Role of Digital Technologies in Deeper Learning,” Students at the Center (2014): 1–22.
[10] US Department of Defense, Ethical Principles for Artificial Intelligence (Washington, DC: Department of Defense, 2020).
[11] John R. Allen and Chris C. Demchak, “Military Organizations and Emerging Technologies,” Defence Studies 21, no. 3 (2021): 345–362.
[12] R. H. Khan et al., “Application of Advanced Simulation Technologies in Military Training,” Journal of Defence Education 12, no. 1 (2022): 55–68.
[13] Sebastian Riedel et al., “Adaptive Simulation Environments for Military Training,” Simulation & Gaming 51, no. 4 (2020): 475–495.
[14] Chris Dede, “The Role of Digital Technologies in Deeper Learning,” Students at the Center (2014): 1–22.
[15] Paul Scharre, Army of None: Autonomous Weapons and the Future of War (New York: W. W. Norton, 2018).
[16] David A. Kolb, Experiential Learning: Experience as the Source of Learning and Development (Englewood Cliffs, NJ: Prentice Hall, 1984).
[17] K. Anders Ericsson, “Deliberate Practice and the Acquisition of Expert Performance,” Psychological Review 100, no. 3 (1993): 363–406.
[18] John Sweller, “Cognitive Load Theory,” Psychology of Learning and Motivation 55 (2011): 37–76.
[19] NATO, NATO Artificial Intelligence Strategy (Brussels: NATO Headquarters, 2021).
[20] Leonard J. Matthews, “The Future of Military Education,” Parameters 49, no. 2 (2019): 5–16.
[21] Matthew B. Caffrey, On Wargaming: How Wargames Have Shaped History and How They May Shape the Future (Newport, RI: Naval War College, 2000).
[22] US Department of Defense, Ethical Principles for Artificial Intelligence (Washington, DC: Department of Defense, 2020).
[23] John R. Allen and Chris C. Demchak, “Military Organizations and Emerging Technologies,” Defence Studies 21, no. 3 (2021): 345–362.
[24] UK Ministry of Defence, Defence Artificial Intelligence Strategy (London: Ministry of Defence, 2022).
[25] J. D. Fletcher and A. P. Wind, Costs, Benefits, and Effectiveness of Military Training (Santa Monica, CA: RAND Corporation, 2013).
[26] Sebastian Riedel et al., “Adaptive Simulation Environments for Military Training,” Simulation & Gaming 51, no. 4 (2020): 475–495.
[27] K. Anders Ericsson, “Deliberate Practice and the Acquisition of Expert Performance,” Psychological Review 100, no. 3 (1993): 363–406.
[28] US Department of Defense, Ethical Principles for Artificial Intelligence (Washington, DC: Department of Defense, 2020).
[29] Chris Dede, “The Role of Digital Technologies in Deeper Learning,” Students at the Center (2014): 1–22.
[30] NATO, NATO Artificial Intelligence Strategy (Brussels: NATO Headquarters, 2021).
[31] Matthew B. Caffrey, On Wargaming: How Wargames Have Shaped History and How They May Shape the Future (Newport, RI: Naval War College, 2000).
[32] NATO Science and Technology Organization, Human–Machine Teaming in Military Operations (Brussels: NATO Science and Technology Organization, 2023).
[33] US Department of Defense, Ethical Principles for Artificial Intelligence (Washington, DC: Department of Defense, 2020).
[34] UK Ministry of Defence, Defence Artificial Intelligence Strategy (London: Ministry of Defence, 2022).
[35] K. Anders Ericsson, “Deliberate Practice and the Acquisition of Expert Performance,” Psychological Review 100, no. 3 (1993): 363–406.
[36] J. D. Fletcher and A. P. Wind, Costs, Benefits, and Effectiveness of Military Training (Santa Monica, CA: RAND Corporation, 2013).
[37] David A. Kolb, Experiential Learning: Experience as the Source of Learning and Development (Englewood Cliffs, NJ: Prentice Hall, 1984).
[38] NATO, Data Exploitation Framework for Defence (Brussels: NATO Headquarters, 2022).
[39] UK Ministry of Defence, Defence Artificial Intelligence Strategy (London: Ministry of Defence, 2022).
[40] Leonard J. Matthews, “The Future of Military Education,” Parameters 49, no. 2 (2019): 5–16.
[41] J. D. Fletcher and A. P. Wind, Costs, Benefits, and Effectiveness of Military Training (Santa Monica, CA: RAND Corporation, 2013).
[42] Chris Dede, “The Role of Digital Technologies in Deeper Learning,” Students at the Center (2014): 1–22.
[43] NATO Science and Technology Organization, Human–Machine Teaming in Military Operations (Brussels: NATO Science and Technology Organization, 2023).
[44] Paul Scharre, Army of None: Autonomous Weapons and the Future of War (New York: W. W. Norton, 2018).
