“AI does not change the objectives of warfare it compresses the timelines, scales the operations, and removes the human friction that has historically been the primary governor of escalation.” AI in warfare refers to the integration of artificial intelligence across military domains targeting, intelligence analysis, logistics, cyber operations, command and control, and strategic decision support transforming the speed, scale, and nature of military operations.
Executive Summary
Military AI integration has moved from research programs to active operational deployment across all major military powers. The United States’ JAIC (Joint Artificial Intelligence Center, now CDAO Chief Digital and AI Office) has deployed AI tools across hundreds of defense use cases. China’s PLA has published AI-enabled military modernization targets for 2027 and 2035. Russia has integrated AI-enhanced drone systems into ongoing combat. Israel’s AI targeting systems have become a focal point for IHL debates. The governance architecture international humanitarian law, rules of engagement, escalation management protocols was designed for human decision-makers operating on human timescales, and is structurally inadequate for AI-augmented or autonomous systems operating in milliseconds.
The Strategic Mechanism
- Intelligence, Surveillance, and Reconnaissance (ISR): AI dramatically expands the volume of sensor data that can be processed in real time, enabling continuous battlefield awareness at scales impossible with human analysts. The US Project Maven program used AI to process drone footage at scale; its principles are now embedded in the CDAO’s operational programs.
- Targeting systems: AI-assisted target identification, battle damage assessment, and engagement recommendation systems accelerate the kill chain from sensor to shooter, reducing the time available for human review of individual engagement decisions.
- Cyber operations: AI enables offensive cyber operations at machine speed, identifying vulnerabilities, generating custom malware, and managing distributed attack infrastructure at a scale and pace that overwhelms human defenders.
- Logistics and sustainment: AI optimization of supply chains, predictive maintenance, and strategic lift planning represents among the highest near-term return-on-investment defense applications, with lower risk and controversy than lethal applications.
- Strategic decision support: AI systems providing decision support to senior commanders during crisis management are the most consequential and least-discussed application a system that flags certain options as strategically optimal may effectively constrain human choices in high-pressure situations.
Market & Policy Impact
- US DoD AI/ML investment exceeded $2.5 billion in fiscal year 2023, with the CDAO overseeing over 685 active AI projects across the department.
- China’s PLA Strategic Support Force incorporates AI-enabled cognitive warfare, electronic warfare, and space operations capabilities as explicit modernization priorities in the 14th Five-Year Plan (2021-2025).
- The US-UK-Australia AUKUS Pillar II advanced capabilities program explicitly identifies AI as a top-tier technology priority alongside autonomous undersea vehicles and hypersonic weapons, funding joint AI defense development.
- Israel’s disclosed use of AI targeting systems (Lavender, Gospel, and others) in Gaza operations generated the most consequential IHL debate about AI in warfare since the development of drone strike programs, prompting formal UN inquiries in 2024.
- NATO’s June 2024 summit communique identified AI-enabled hybrid warfare combining cyber operations, autonomous systems, and disinformation as the primary near-term threat vector facing the alliance.
Modern Case Study: Project Maven and the Pentagon’s AI Doctrine, 2017-2025
Project Maven was the US Department of Defense’s first large-scale operational AI program, launched in 2017 to use computer vision to analyze drone footage from combat zones reducing the analyst workload of reviewing hundreds of hours of video daily. The project triggered a significant internal debate at Google, whose employees’ protests led to the company not renewing its contract in 2018. Maven’s underlying mission transferred to Palantir and eventually into the CDAO. By 2024, Maven had evolved into Maven Smart System, providing AI-enhanced targeting and intelligence analysis for operational use. The program’s trajectory from research experiment to operational infrastructure mirrors AI adoption across the Pentagon more broadly. It also established the pattern of contractor-government AI partnerships that have defined US military AI acquisition, and generated the first documented employee ethics revolt over military AI a governance challenge that has since spread to Amazon, Microsoft, and other major defense contractors.