Rapid-fire AI Abilities: Exploring Capacities and Constraints of Artificial Intelligence
The United States Army is advancing its targeting methodology by integrating artificial intelligence (AI), a move that promises to accelerate targeting cycles while preserving accountability through human oversight.
This Army doctrine, known as D3A (Decide, Detect, Deliver, Assess), integrates multiple warfighting functions across the targeting cycle to enable precision effects on adversaries. The necessity of this integration is paramount for the Army, as AI excels at scaling analysis speed, processing vast intelligence, surveillance, and reconnaissance (ISR) data to identify targets faster than human operators alone.
However, the use of AI in targeting raises ethical concerns. Accordingly, military doctrine emphasizes AI must remain a tool supporting commanders, not supplanting their control to avoid misuse or errors in lethal targeting decisions. The Army continues to rely on D3A as the doctrinal cornerstone of fires and effects integration at the brigade and division levels.
Incorporating AI into D3A is about optimizing the targeting workflow, accelerating sensor-to-shooter kill chains, reducing cognitive burden, and improving commanders' decision-making in contested environments. Optimization algorithms and prescriptive analytics can refine weapon-target pairing and target engagement timings for the deliver phase of D3A. Tools like game theory models, decision trees, logistic regression algorithms, and clustering models can support enemy course-of-action development, attack asset prioritization, and effects determination in the decide phase of D3A.
AI technologies have proven utility in intelligence, surveillance, and reconnaissance processing, decision support, and autonomous systems operations. However, large language models (LLMs), such as Meta's LLaMA, do not inherently grasp doctrinal terminology or contextual nuance. A modular and doctrinally grounded approach is required to adapt AI to the D3A methodology.
The Army's targeting methodology, D3A, is described in Field Manual 3-60, Army Targeting. The necessity of integrating AI into this methodology is evident, as AI offers undeniable scaling advantages, particularly in data processing and decision acceleration. However, it is critical that AI functions as an augmentation tool rather than replacing human judgment. Human commanders must remain the final decision-makers—exercising "human-on-the-loop" oversight—to ensure ethical considerations such as rules of engagement, proportionality, and military necessity are properly validated before lethal force is authorized.
In the assess phase of D3A, battle damage estimation benefits from clustering models and explainable AI tools. AI-enabled targeting doctrine must codify decision points where human intervention is not just preferred—but required. Time is the most compelling performance metric for evaluating AI effectiveness in the targeting process.
Jesse R. Crifasi, a retired US Army chief warrant officer 4, a senior advisor in the defense industry specializing in joint fires and targeting, and a PhD student in public policy and national security at Liberty University, has authored multiple doctrinal and technical assessments on digital fires and artificial intelligence integration in targeting operations. Crifasi's views expressed in this article do not reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense.
The image used in this article was credited to Sgt. Rebecca Watkins, US Army. The article was published.
[1] Field Manual 3-60, Army Targeting
1) The United States Army's doctrine for targeting, D3A, is being enhanced by integrating artificial intelligence (AI), aiming to speed up targeting cycles while maintaining accountability through human oversight.2) AI excels at analyzing vast amounts of intelligence, surveillance, and reconnaissance (ISR) data to identify potential targets more quickly than human operators alone.3) AI functions as a tool to support commanders in the targeting process, rather than supplanting their control, to avoid misuse or errors in lethal targeting decisions.4) The effective use of AI in the Army's targeting methodology requires a doctrinally grounded approach to adapt AI to the D3A methodology, ensuring its utility in warfighting functions across the targeting cycle.
