The official described it as an example of how things might work but wouldn’t confirm or deny whether it represents how AI systems are currently being used.
Other outlets have reported that the Anthropics Cloud has been integrated into existing military AI systems and used in operations. Iran And Venezuelabut the official’s comments add insight into the specific role that chatbots can play, particularly in speeding up target searches. He also highlighted how the military is deploying two different AI technologies, each with distinct limitations.
Since at least 2017, the US military has been working on a “big data” initiative called Maven. It uses older types of AI, specifically computer vision, to analyze seas of data and imagery collected by the Pentagon. Maven can take thousands of hours of aerial drone footage, for example, and algorithmically identify targets. A 2024 Report from Georgetown University showed soldiers using the system to select and evaluate targets, speeding up the process of getting those targets approved. Soldiers interacted with Maven through an interface with a battlefield map and dashboard, which could highlight potential targets in one color and friendly forces in another.
Creative AI is now being added as a conversational chatbot layer—which the military can use to more quickly search and analyze data as it decides which targets to prioritize, the official’s comments suggest.
Generative AI systems, such as those that underpin ChatGPT, Claude, and Grok, are fundamentally different technologies from the AI that primarily powers Maven. Built on large tongue models, they are much less battle-tested. And while Maven’s interface allowed users to directly inspect and interpret the data on the map, the output produced by generative AI models is easy to access but difficult to verify.
Using generative AI for such decisions is reducing the time needed in the targeting process, added the official, who did not provide details when asked how much additional speed is possible if humans need to spend time double-checking model outputs.
Following the recent strike on a girls’ school in Iran that killed more than 100 children, public scrutiny of the use of military AI systems has increased. Multiple News Outlets reported that the attack was carried out by a US missile, although the Pentagon said it was under investigation. And while The Washington Post is Reported While Claude and Maun have been involved in targeting decisions in Iran, there is still no evidence to explain what role, if any, generative AI systems played. gave The New York Times Reported On Wednesday, a preliminary investigation found outdated targeting data to be partially responsible for the strike.