AI Targeting System Guided Iran Strike Planning

Artificial intelligence systems played a central role in the planning of recent U.S. and Israeli strikes on Iran, according to a report by The Washington Post citing three sources familiar with the matter. The report describes how the Pentagon integrated Anthropic’s Claude artificial intelligence model into Palantir’s Maven Smart System, allowing military planners to generate and prioritize large numbers of potential targets across Iran.

The system was used to support critical military planning tasks including real-time target identification, live surveillance analysis, and the processing of incoming intelligence. According to the report, the Maven platform used Claude’s AI capabilities to identify possible targets, determine their precise geographic coordinates, and rank them according to operational importance.

Two sources told the newspaper that while planning a potential strike campaign against Iran, the AI-assisted system produced hundreds of proposed targets and supplied exact location coordinates for them. The system then sorted these sites by priority level, enabling U.S. and Israeli planners to refine operational targeting lists.

According to the report, U.S. planners used the AI-enabled system extensively during the opening phase of the military operation. The Pentagon allegedly relied on the technology to help identify and plan strikes on approximately 1,000 targets during the first day of attacks against Iran.

The military campaign began on February 28, when Israel and the United States launched coordinated strikes while negotiations between Tehran and Washington were still underway. Iran subsequently responded with attacks targeting locations connected to U.S. forces in the region, including sites in Qatar, the United Arab Emirates (UAE), and Bahrain.

Sources cited by The Washington Post said the Pentagon had begun integrating the Claude chatbot into the Maven system in late 2024. The AI tool was used to generate recommended target lists, track logistical requirements, and produce summaries of intelligence collected from the field.

Pentagon–Anthropic Contract Tensions

The report also describes tensions between the U.S. Department of Defense and Anthropic over how the company’s technology could be used in military operations. According to the article, U.S. Defense Secretary Pete Hegseth issued an ultimatum on February 24, warning the company that it risked losing government contracts unless the Pentagon was allowed to use the AI system more freely by February 27.

The dispute escalated shortly afterward. On February 27, U.S. President Donald Trump ordered federal agencies to immediately stop using Anthropic’s technology following disagreements between the Pentagon and the company over access and operational control.

Despite the dispute, the Maven program continued expanding within the U.S. military. By May 2025, more than 20,000 military personnel were using the system across multiple branches of the armed forces. The expansion reflected the Pentagon’s broader push to integrate artificial intelligence into everyday planning and intelligence workflows.

One source also told the newspaper that the Claude system had previously been used in other sensitive operations. According to the source, the AI tool was involved in planning efforts connected to an operation that brought Venezuelan President Nicolas Maduro to the United States through a military action.

The growing adoption of the Maven platform underscores how AI tools are becoming embedded in operational military decision-making. The integration of Claude into Maven created a system capable of scanning intelligence feeds, generating target options, assigning priority levels, and presenting planners with structured strike lists that could be executed during the early stages of a conflict.