What 2025 Taught Us About AI Adoption in the DoW—and What 2026 Will Demand

In 2025, the U.S. Department of War and military services crossed a threshold in their use of artificial intelligence. New, faster acquisition pathways came online. GenAI.mil moved from concept to reality. Early results from AI-enabled Navy shipbuilding signaled real operational promise. Across the Department, pilots, prototypes and proofs of concept multiplied.

The question is no longer whether the DoW will adopt AI. AI is a reality in the military. The question now is whether the Department is prepared to operationalize AI at scale—under real-world conditions, under pressure and at the speed of conflict.

A review of 2025 research from McKinsey, Boston Consulting Group, Microsoft, Accenture, the World Economic Forum and leading academic institutions points to a single, consistent conclusion:

AI adoption is no longer constrained by access to technology. It is constrained by force readiness, skills and institutional adaptation.

For the DoW, AI is also not a technology issue. The Department already has—and will continue to have—access to the most advanced AI capabilities in the world. The decisive question is whether its workforce is prepared to use them effectively.

What follows are what I believe are the most relevant insights from the 2025 research to the DoW—and what they mean for 2026 and beyond.

AI is spreading fast but decisive advantage remains uneven.

By 2025, AI tools were commonplace across most industry sectors. Yet widespread use did not translate into consistent impact.

McKinsey’s State of AI 2025 report communicated broad deployment but only a small share of organizations achieving sustained results. BCG’s AI at Work 2025 report found high tool usage but limited performance gains when workflows, authorities and decision rights remain unchanged. Microsoft’s 2025 Work Trend Index reported that technology adoption now far outpaces organizational integration.

The DoW does not lack pilots, sandboxes or demonstrations. The Department lacks operational absorption of AI.

Until AI becomes embedded in command and control, intelligence analysis, logistics and sustainment, training and readiness generation and operational planning, it remains peripheral. Decisive advantage comes not from owning AI tools, but from integrating them into how the military actually fights and operates.

Workforce readiness is now the primary constraint to AI adoption.

cross all major 2025 studies, one finding dominated: workforce readiness has overtaken technology as the primary barrier to AI adoption.

McKinsey identified talent, leadership capability and change management—not algorithms—as the most significant barriers to scale. BCG found dramatically higher performance gains among employees who receive formal AI training. Kyndryl’s People Readiness Report 2025 communicated that AI “pacesetters” invest in broad workforce literacy, not narrow expertise. Pluralsight’s AI Skills Report 2025 revealed a persistent disconnect between how capable workers perceive themselves to be and how capable they actually are.

For the DoW, the implication is straightforward. AI systems only perform as well as the commanders, officers, warfighters, analysts and other personnel who employ them.

In the military AI adoption is not a technical gap. It is a development gap.

AI literacy—not specialization—is the decisive enabler.

One of the most important lessons of 2025 is that AI advantage does not come from creating more engineers.

High-performing organizations prioritize AI and data literacy among their leaders and employees They train personnel to understand uncertainty, bias and failure modes. They emphasize human–AI teaming in decision making. They redesign tasks and processes instead of blindly automating legacy workflows.

Microsoft research showed that organizations with higher confidence in AI adopt it more deeply across missions. World Economic Forum data indicated that applied AI skills—not deep technical specialization—represent the fastest-growing demand. Academic research from INSEAD found that generative AI shifts value toward higher-order human judgment rather than away from it.

For the DoW, the message is clear. Decisive advantage comes from AI-capable leaders and warfighters at scale—not from isolated technical elites.

Training alone will not produce AI adoption.

The 2025 research also warned that AI adoption efforts that includes training but lack institutional change will fail.

Accenture’s Technology Vision 2025 report emphasized the redesign of roles, authorities and workflows. McKinsey’s work on “superagency” highlighted the importance of permission and trust, not just skill acquisition. WalkMe’s Digital Adoption 2025 report communicated that when digital tools are not embedded into daily processes, sustained use collapses.

For the DoW, AI upskilling must coincide with deeper change. Doctrine and concept of operations must evolve. Authorities for AI-supported decisions must be explicit. Accountability frameworks for human–AI teaming must be clear. Leaders must build institutional trust in AI outputs under operational pressure.

Absent these changes, even well-trained personnel will revert to legacy decision processes when the stakes rise.

AI-ready military commands treat upskilling as a warfighting capability.

Absent these changes, even well-trained personnel will revert to legacy decision processes when the stakes rise.

What does this mean for 2026?

The evidence from 2025 points clearly to what comes next.

First, AI readiness will become a measurable element of DoW readiness. Expect greater emphasis on AI-enabled roles, greater confidence in AI-assisted decision making and improved operational outcomes enabled by AI.

Second, AI literacy will become a command expectation. Commanders and senior leaders will be expected to understand the capabilities and limits of AI, integrate them into planning and manage risk responsibly.

Third, DoW workforce development, doctrine and technologies will converge. AI will force tighter alignment among training pipelines, acquisition and software delivery and operational design and command authority.

Fourth, mid-level leaders will determine AI adoption success or failure. As in civilian organizations, AI adoption hinges on captains, majors, senior non-commissioned officers and GS-13/14 civilian personnel—the leaders who translate intent into action.

Finally, readiness gaps will widen. DoW organizations that delay workforce upskilling will fall behind—even if they retain equal access to advanced AI tools.

Closing thought

The message of 2025 is unmistakable.

AI advantage is not about possessing advanced algorithms. It is about preparing the DoW to fight, decide, and adapt differently.

In 2026, the Department will no longer be judged by experimentation. It will be judged by operational trust, institutional readiness and the ability to translate AI into a decisive advantage on the battlefield.

Dr. J. Keith Dunbar
Founder and Chief Executive Officer
FedLearn

References

McKinsey — “Superagency in the workplace: Empowering people to unlock AI’s full potential” (Jan 28, 2025) Emphasizes people/leadership and capability building as key constraints on scaling AI in the workplace. McKinsey & Company
McKinsey — “The State of AI: Global Survey 2025” (2025 edition) Annual survey on how organizations are “rewiring” to capture AI value, includes operating model and talent elements tied to adoption outcomes. McKinsey & Company+1
BCG — “AI at Work 2025: Momentum Builds, but Gaps Remain” (Jun 26, 2025) Calls out training/upskilling and workflow redesign as key to moving from tool rollout to real value. BCG Global+1
Microsoft — “2025 Work Trend Index Annual Report” (fielded Feb–Mar 2025) Large cross-country study (survey + labor market + productivity signals) on AI at work. Discusses capability gaps and organizational readiness affecting adoption. Azure CDN
World Economic Forum — “New Economy Skills: Building AI, Data and Digital Capabilities for Growth” (2025) Directly focused on AI/data/digital capabilities (supply/demand + credentialing) for competitiveness—useful for upskilling strategy tied to adoption. World Economic Forum Reports
Kyndryl — “People Readiness Report 2025” (2025) Measures workforce readiness to leverage AI. Highlights skills gaps as a barrier and contrasts “pacesetters” vs others on adoption enablers. Kyndryl
Pluralsight — “AI Skills Report 2025” (2025) Survey-based view of perceived vs actual AI skills gaps and how those gaps block organizational AI adoption. Pluralsight
WalkMe — “The State of Digital Adoption 2025: AI edition” (2025) Digital adoption lens on AI. Pairs survey + usage/adoption data. Discusses workforce enablement as part of adoption success. WalkMe – Digital Adoption Platform
Wharton / GBK Collective — “2025 AI Adoption Report: Gen AI Fast-Tracks Into the Enterprise” (Oct 28, 2025) Year-over-year executive study. Explicitly includes “training” and people/process levers tied to enterprise adoption and ROI. Knowledge at Wharton
Accenture — “Technology Vision 2025: AI: A Declaration of Autonomy” (2025) Strategy/trends report that repeatedly stresses capability building (AI fluency) as organizations shift from pilots to reinvention. Accenture
KPMG — “AI and GBS: Driving value and evolving the workforce” (2025) Focuses on GenAI in Global Business Services and the workforce changes/skills needed to realize value at scale. KPMG
SEI (Carnegie Mellon) + Accenture — AI Adoption Maturity Model (announcement + paper context, 2025) Maturity model work explicitly about organizational capability and readiness for AI adoption (useful for linking skills/readiness to adoption maturity). CMU School of Education
Springer (AI and Ethics) — “A review of global reskilling and upskilling initiatives in the age of AI” (Published Jun 23, 2025) Systematic review (2024–2025 literature) centered on upskilling/reskilling initiatives and AI literacy—useful grounding for the “skills → adoption” argument. Springe14.
arXiv working paper (INSEAD) — “Generative AI Adoption and Higher Order Skills” (Mar 2025) Uses firm job postings to analyze how GenAI adoption shifts skill demand—evidence for the skills side of adoption at scale. arXiv