AI First Defense

Military AI-First Strategy

The January 2026 Acceleration Directive

On January 9, 2026, the U.S. Department of War issued a sweeping memorandum titled "Artificial Intelligence Strategy for the Department of War," directing the military establishment to become an "AI-first" warfighting force across all components. The directive, followed by Secretary of Defense Pete Hegseth's January 12 speech, set an explicit agenda for AI acceleration at what the Department described as "wartime speed." This represented a decisive shift from earlier AI strategies issued between 2018 and 2023, which had focused primarily on governance frameworks, infrastructure planning, and responsible AI principles. The 2026 strategy instead prioritized operational execution, bureaucratic barrier removal, and measurable deployment timelines.

Pace-Setting Projects and Institutional Architecture

Central to the AI-first military posture are seven Pace-Setting Projects (PSPs) administered by the Chief Digital and Artificial Intelligence Office (CDAO). These projects span three categories of military capability: combat operations, intelligence pipelines, and enterprise systems. Among them, Swarm Forge establishes a competitive mechanism for discovering and scaling novel approaches to fighting with and against AI-enabled capabilities. Each PSP operates under a single accountable leader with aggressive timelines and monthly progress demonstrations to the Deputy Secretary of War and Under Secretary for Research and Engineering. Initial demonstrations were scheduled for July 2026, six months from the memorandum's issuance.

The institutional architecture supporting this AI-first posture extends well beyond the PSPs themselves. The strategy designates the Under Secretary of War for Research and Engineering as the single Chief Technology Officer, creates a CTO Action Group empowered to clear procurement and authorization blockers, and elevates the Defense Innovation Unit (DIU) and Strategic Capabilities Office (SCO) as core innovation organizations. A monthly Barrier Removal Board was established to compress contracting, testing, and authorization timelines that traditionally stretched across years of acquisition cycles.

Data Access, Compute Infrastructure, and Talent

Three foundational enablers underpin the AI-first military transformation. First, the strategy mandates strict enforcement of the DoD Data Decrees, requiring military departments and components to deliver federated data catalogs to the CDAO within 30 days. The CDAO received authority to direct release of any Department data to cleared users with a valid purpose, with denials required to be justified within seven days. Second, the directive calls for substantial expansion of AI compute infrastructure from centralized data centers to tactical edge environments where warfighters operate. Third, accelerated AI talent acquisition programs using special hiring and pay authorities became mandatory across all components, with each branch required to submit AI hiring plans within 60 days.

The strategy also imposed procurement requirements with significant implications for the defense industrial base. AI system architectures must conform to Modular Open Systems Architecture (MOSA) standards enabling component replacement at commercial velocity. Most notably, the CDAO must ensure AI vendors can deploy their latest models within 30 days of public release, a cadence closer to commercial software deployment than anything previously seen in defense acquisition. Five of six military branches had already elevated GenAI.mil as their enterprise AI platform by the time the strategy was issued.

International Context and Allied Adoption

The U.S. AI-first military posture exists within a broader international context. NATO allies have undertaken parallel efforts to integrate AI into defense planning, though at varying speeds and scales. The United Kingdom's Defence AI Centre, established to coordinate AI adoption across the Ministry of Defence, has focused on autonomous logistics and intelligence analysis. Australia's Advanced Strategic Capabilities Accelerator (ASCA) targets AI-enabled undersea warfare and autonomous systems. France's Agence de l'Innovation de Defense has invested in AI for predictive maintenance and satellite imagery analysis. These programs collectively demonstrate that "AI-first defense" as a concept transcends any single national strategy, reflecting a broader doctrinal evolution in how modern militaries conceptualize technological advantage.

Cybersecurity AI-First Defense

The Platform Consolidation Paradigm

In enterprise cybersecurity, "AI-first defense" describes an architectural approach where machine learning and AI-driven analysis serve as the primary detection and response layer rather than supplementing traditional signature-based tools. This paradigm has driven a wave of platform consolidation as major vendors build unified security architectures with AI at their core. CrowdStrike's Falcon platform, which the company projects will address a total addressable market reaching $300 billion by 2030, exemplifies this approach by using AI-native threat detection across endpoints, cloud workloads, and identity systems. SentinelOne's Singularity platform takes a similar AI-first approach, with the company reporting 23% quarterly revenue growth in 2025 and projecting annual recurring revenue surpassing $1 billion in early 2026.

The market dynamics supporting AI-first cybersecurity are substantial. Global security spending reached approximately $236 billion in 2025, with AI-driven cybersecurity specifically projected to grow at a 27.8% compound annual growth rate to reach an estimated $82 billion by 2029. This growth is propelled by the accelerating sophistication of AI-powered attacks: phishing attacks surged over 1,200% following the widespread availability of generative AI tools, and ransomware incidents increased 84% year-over-year. Traditional signature-based and rule-driven defenses cannot match the speed and adaptability of these threats, creating structural demand for AI-native defensive architectures.

AI Detection and Response as Operational Doctrine

The operational doctrine of AI-first cybersecurity centers on continuous behavioral analysis rather than periodic scanning. CrowdStrike's September 2025 acquisition of AI security specialist Pangea for a reported $260 million aimed to deliver what the company described as the first complete AI Detection and Response (AIDR) service. SentinelOne's acquisition of Observo AI for $225 million in the same period added AI-native telemetry pipeline management to its existing platform. Check Point's agreement to acquire Lakera brought AI-native security for agentic AI applications into its stack. These acquisitions collectively signal that the major cybersecurity vendors are restructuring their platforms around AI as the foundational detection and response architecture.

Fortinet expanded its hardware portfolio with the FortiGate 3800G, an appliance powered by custom NP7 and SP5 chips capable of scanning network traffic for threats at 800 gigabits per second. SentinelOne's Purple AI assistant automates breach investigation and detection rule generation, while CrowdStrike introduced Charlotte Agentic SOAR for creating cybersecurity automation workflows using multiple AI agents. The common thread across these developments is the treatment of AI not as an enhancement to existing security operations but as the primary operational architecture around which human analysts, automated workflows, and hardware infrastructure are organized.

AI Governance as the Next Security Frontier

As organizations accelerate AI adoption, a parallel challenge has emerged: securing the AI systems themselves. JetStream Security, founded by veterans of CrowdStrike and SentinelOne, raised $34 million in seed funding in early 2026 from Redpoint Ventures with participation from CrowdStrike CEO George Kurtz and Wiz CEO Assaf Rappaport. The company's AI Blueprints feature creates real-time graphs mapping everything an AI system does inside an organization, tracing which agents are running, which models they use, and what data they access. This represents an emerging "AI-first defense of AI" paradigm: using AI-native tools to secure the AI systems that now constitute the primary defensive infrastructure. With global AI spending projected to reach $650 billion in 2026, the governance gap between AI deployment and AI security represents both a critical risk and a significant market opportunity.

Biodefense AI Preparedness

Computational Biodefense and Early Warning Systems

The biodefense sector presents a distinct application of AI-first defense principles, where artificial intelligence serves as the primary architecture for pathogen detection, countermeasure development, and outbreak prediction. Traditional biodefense models relied on laboratory-based identification and stockpile-centered response. AI-first approaches instead embed predictive analytics and computational biology at the foundation of preparedness architectures, enabling continuous genomic surveillance, accelerated vaccine candidate identification, and epidemiological modeling that incorporates behavioral and economic feedback loops.

The Biomedical Advanced Research and Development Authority (BARDA), operating under HHS's Administration for Strategic Preparedness and Response (ASPR), has historically anchored U.S. civilian biodefense through programs like Project BioShield and the Strategic National Stockpile. BARDA's FY2026 budget request of $654 million, though reduced from the prior year, continues to fund advanced development of medical countermeasures against chemical, biological, radiological, and nuclear threats. The DoD separately allocated approximately $4 billion for biodefense programs in 2026, with the FY26 Reconciliation Bill making an additional $2 billion available to the Defense Health Program through FY29.

AI-Native Biodefense Startups

Two venture-backed companies that emerged in late 2025 illustrate the AI-first approach to biodefense. Valthos raised $30 million in October 2025 from OpenAI Startup Fund, Founders Fund, and Lux Capital to build what it described as the computational infrastructure for biodefense, using AI methods to characterize pathogen sequences and design adaptive countermeasures. Red Queen Bio, structured as a public benefit corporation, raised $15 million in December 2025 with a focus on what it terms "defensive co-scaling," coupling defensive computational capabilities directly to advances in AI capability. The distinction between these approaches is significant: Valthos functions as a rapid response capability for emerging threats, while Red Queen proactively stress-tests AI systems to identify dangerous outputs and develops countermeasures preemptively.

These startups operate alongside established pharmaceutical companies that announced over $160 billion in U.S. manufacturing reshoring commitments in 2025, partly to address supply chain vulnerabilities exposed during the COVID-19 pandemic. The convergence of AI capability, synthetic biology tools, and geopolitical pressure on pharmaceutical supply chains is creating structural demand for AI-first biodefense capabilities that can operate at the speed of emerging biological threats rather than the traditional multi-year countermeasure development timeline.

Genomic Surveillance and Predictive Epidemiology

AI-first defense in the biodefense context extends to genomic surveillance networks that continuously monitor pathogen evolution across global populations. Johns Hopkins researchers have developed feedback-informed epidemiological models that blend economic decision-making with disease dynamics, capturing how individual behavioral choices affect outbreak trajectories. These models represent a departure from static epidemiological forecasting toward adaptive systems that learn from real-world data in near real-time. The Council on Strategic Risks has expanded its Mid-Career Biodefense Bootcamp to include professionals from organizations spanning ARPA-H, BARDA, Google DeepMind, and the Hoover Institution, reflecting the increasingly interdisciplinary nature of biodefense that bridges AI research, public health policy, and national security strategy.

Key Resources

• Artificial Intelligence Strategy for the Department of War -- Secretary of Defense Memorandum, January 9, 2026 (PDF)
• U.S. Biodefense Budget Breakdown: Fiscal Year 2026 Update -- Council on Strategic Risks
• AI Cybersecurity Trends to Watch in 2026 -- SentinelOne Research
• The U.S. DoD's AI Acceleration Strategy -- Defense Security Monitor, Forecast International
• Administration for Strategic Preparedness and Response (ASPR) -- U.S. Department of Health and Human Services