Washington supports Riyadh with targeting algorithms in the Yemen war, deploying two hundred U.S. soldiers and a joint intelligence room

In Joint Command and Control Rooms, military superiority today is no longer measured by the size of munitions but is decided by how quickly an intelligence picture can be processed before the launch platform leaves its mountain position. As military tension resurfaces on Saudi Arabia's southern border and the Houthi group escalates attacks on airports and oil facilities in September, Washington chose to support Riyadh through an advanced digital pathway. After U.S. President Donald Trump rejected the option of direct military intervention at the request of the Saudi leadership, the U.S. Department of Defense dispatched between one hundred and two hundred personnel as part of a new joint forces command, aiming to provide Saudi army officers with an AI-enabled targeting tool to monitor the battlefield moment by moment.

This technology is based on an architecture similar to the Pentagon’s smart Maven system, which was developed in a project that began in 2017 and was field-tested to strike targets in Iran, Iraq and Syria. The platform, which the former Google product manager Bellaul Sidho described as akin to a Google Earth for war, performs a task that was almost impossible for human analysts: it collects, classifies and streams visual data from hundreds of direct satellites and drone cameras, then detects anomalous patterns and pinpointed missile launchers, tanks and suspicious vehicle convoys with precision, while calculating the optimal angle for a strike. This algorithmic integration shortens what military doctrine calls the “kill chain,” the interval between detecting a threat and issuing a bombardment order. The effect was evident in the U.S. attack on Iran in late February, when the platform raised processing capacity from roughly a hundred targets per day to enabling the strike of more than a thousand targets within the first twenty-four hours.

But the rugged terrain in Yemen imposes a calculus that differs fundamentally from conventional battlefields. Houthi fighters rely on a hit-and-run style using trucks that carry ballistic-missile platforms and launch swarms of small drones from among hills and deep valleys, and they often vacate their positions before the smart device finishes transmitting its coordinates. In addition, the system’s algorithms, designed to operate efficiently in open, sparsely populated environments, run into environmental factors that reduce their accuracy such as desert dust storms, strong winds, deliberate camouflage and terrain changes. This military dilemma is compounded by the ideological nature of the opponent; Faar al-Muslami, a researcher at the Chatham House institute, notes that the group adopts a zero-sum approach that views suffering and poverty as a doctrinal trial, which gives it an exceptional ability to absorb material destruction and remain on the field.

The problem exceeds the limits of tactical efficiency and collides with a high human and moral cost arising from accelerated targeting decisions. Although military protocol assumes a human review to verify every target proposed by the algorithm, time pressure often leads to dropping safety margins, as documented by the targeting of a primary school in southern Iran on the first day of operations with the Maven system, an Iranian targeting platform, which resulted in the death of nearly two hundred people, most of them children aged seven to twelve. Similar concerns were seen in Gaza under the Israeli Hobscura system, an Israeli targeting platform, where researchers Frederick Verry and Andrew Boony, together with writer Anthony Lewinstein, note that lowering human verification standards increases civilian casualties without achieving a decisive military outcome against armed groups.

This shift places operations leaders and defensive planning officials in the Gulf region before the fact that technical superiority does not compensate for shortcomings in field and political understanding.The shift toward automating the operations theatre transfers the strategic weight from purchasing traditional missile-intercept systems to the necessity of building an integrated sovereign architecture for locally collecting and analyzing intelligence data. Arab armies cannot rely solely on importing foreign algorithms designed for different environments; rather, there is a urgent need to train national personnel capable of scrutinizing digital outputs and exercising strict human veto power before pulling the trigger, to avoid the disasters of inaccurate targeting and to prevent sliding into an algorithmic arms race that raises the cost of war without determining its outcome.