International Reports: AI Leads a Revolution in US War Management and Target Generation

International press reports have revealed a radical shift in the US military's war management strategies, with artificial intelligence becoming the primary driver of battlefield decision-making processes. This development was clearly evident during recent military operations in Iran, where the Pentagon was able to target over two thousand sites in just four days, a pace that far exceeds traditional human capabilities in planning and analysis.

This unprecedented pace relies on advanced AI systems that process massive streams of intelligence data derived from satellites and drones. These technologies allow for the immediate generation of bombing options, reducing the time needed to identify targets from days and hours to mere seconds or minutes. This marks the first widespread field use of generative AI models in battlefields.

Informed sources clarified that over the past two years, the US Department of Defense has extensively integrated AI-powered technologies, foremost among them the 'Maven Smart System'. Developed by 'Palantir', this system serves as a core operating platform for Pentagon data and is used alongside advanced language models like 'Claude' to analyze combat situations in real-time.

Technical experts believe that the great importance of these models lies in their transition from merely summarizing data to the stage of inference and logical step-by-step reasoning in military problems. This shift has led to a massive increase in the volume of military decisions made, giving field commanders superior maneuverability and rapid response capabilities in complex, information-dense combat environments.

Despite the technological promises of accelerating military decisive action, these technologies raise deep concerns regarding oversight and legal accountability. Debate has recently escalated over the limits of using these models after disagreements between technology companies and the Pentagon, especially concerning the risks associated with generating targets that may not undergo sufficient human review before execution.

Human rights reports highlighted the repercussions of this extreme speed, pointing to the bombing of sensitive civilian facilities such as a girls' primary school in the Iranian city of Minab. Ambiguity still surrounds the extent of AI systems' involvement in including such sites in targeting lists, and whether the error resulted from a technical malfunction or inaccurate human assessment.

Data from the Iranian Red Crescent indicates that joint operations by the United States and the occupying state resulted in damage to over 20,000 non-military buildings, including thousands of residential units. This massive scale of destruction raises urgent questions about the criteria for target selection amidst increasing reliance on algorithms that may lack human sensibility and ethical judgment.

Researcher Jessica Dorsey compared the current campaign with the previous campaign against ISIS, illustrating the vast difference in operational efficiency and speed. While the international coalition took six months to carry out two thousand strikes in Iraq and Syria, US forces achieved the same number in just four days, highlighting the extent of the transformation technology has brought to the 'kill chain'.

The 'Maven' system acts as the software brain that manages the entire kill chain, from target identification and prioritization to weapon selection and damage assessment. This chain previously relied on bureaucratic paper procedures requiring approvals from senior leadership, which significantly delayed military operations.

Sophia Goodfriend, a researcher at Cambridge University, confirms that large language models have demonstrated a superior ability to prepare massive target lists compared to human effort. This technology allows armies to operate at an unprecedented scale, making aerial targeting a continuous and comprehensive process covering vast geographical areas in record time.

According to statements from US geospatial intelligence officials, the number of 'Maven' system users exceeded 20,000 across dozens of military entities by mid-2025. Estimates indicate that this number is continuously rising, with international allies such as NATO joining to use this advanced system in their operations.

US military leaders aim to achieve an ambitious goal of making a thousand high-quality decisions in just one hour on the battlefield. This approach reflects the desire to transform war into a precise computational process, where targets are excluded or selected based on instantaneous analyses performed by machines at lightning speed.

In addition to targeting systems, AI is used in autonomous navigation and computer vision in various conflict zones, including Gaza and Ukraine. Image recognition programs help identify missile launchers and mobile military assets, solving the 'bottleneck' problem soldiers faced when manually reviewing drone footage.

Dorsey concludes her questions about the extent to which actual human control can be exercised over systems that perform millions of calculations per second. The biggest challenge for international law remains how to track these decisions and hold those responsible accountable in the event of violations, especially when 'human judgment' becomes a mere formality in the face of machine speed.

Artificial intelligence has allowed for a qualitative leap in the volume of decisions and the speed with which military personnel can make those decisions during complex operations.