Artificial Intelligence Strategy for the Department of War and the Institutionalization of an AI-First Military
The January 9, 2026 memorandum titled Artificial Intelligence Strategy for the Department of War—issued by the Office of the Secretary of War to senior Pentagon leadership, combatant commanders, and defense agency leaders—sets out an explicitly acceleration-focused doctrine for military AI adoption and organizational redesign. The document frames AI not as a discrete technology program, but as a strategic determinant of military advantage, arguing that AI-enabled warfare and capability development will reshape the character of military affairs over the next decade. It also links this posture to Executive Order 14179 and positions the Department’s objective as becoming an “AI-first” warfighting force across all components.
The memorandum’s central contribution is its operationalization of AI strategy through implementation mechanics rather than abstract policy language. It identifies four immediate priorities: expanding experimentation with leading U.S. AI models, removing bureaucratic barriers to integration, concentrating investment around U.S. advantages in compute and innovation, and executing “Pace-Setting Projects” (PSPs) to build foundational enablers such as infrastructure, data, models, policy, and talent. This is an important shift because it treats institutional friction—not merely technical maturity—as the primary obstacle to AI adoption.
The strategy’s seven PSPs further demonstrate this implementation-oriented approach by dividing action across warfighting, intelligence, and enterprise functions. Programs such as Swarm Forge, Agent Network, and Ender’s Foundry focus on operational experimentation and simulation, while Open Arsenal and Project Grant target intelligence-to-capability conversion and deterrence analysis. On the enterprise side, GenAI.mil and Enterprise Agents seek to normalize AI use across the Department’s civilian and military workforce. The memo also imposes accountability architecture: single responsible leaders, aggressive timelines, measurable outcomes, and monthly reporting to senior officials.
Equally significant is the memo’s emphasis on enablers and governance. It directs expanded AI compute investments, stronger data accessibility through federated catalogs and CDAO enforcement authority, and accelerated hiring using special authorities. It also establishes an execution culture defined by speed, rapid model refresh cycles, and aggressive removal of blockers, including ATOs and other approval bottlenecks. Notably, the memorandum requires model deployment cadences that can support updates within 30 days of public release, signaling a procurement and integration tempo closer to commercial software than traditional defense acquisition timelines.
From an institutional perspective, the memorandum is authored as a transformation directive rather than a technical roadmap. Its final section makes clear that becoming “AI-first” requires reimagining legacy workflows, TTPs, and operational concepts as if modern AI had existed when they were designed. That framing is analytically important: it moves the discussion from AI insertion to organizational redesign, with 2026 cast as a benchmark year for raising the Department’s “AI fitness standards.” As a policy text, it should therefore be read as both a strategy memorandum and a management doctrine for accelerated defense adaptation.
Credit to authors/source: This summary is based on the memorandum Artificial Intelligence Strategy for the Department of War (dated January 9, 2026), issued by the Office of the Secretary of War and addressed to senior Department leadership.
Disclaimer: This blog post is an informational summary and analysis for educational purposes only. It is not legal advice, policy advice, or a definitive interpretation of U.S. government strategy. Readers should consult the original memorandum and qualified counsel or subject-matter experts before relying on this content for compliance, procurement, or operational decisions.