The U.S. Air Force is planning to use artificial intelligence algorithms to manage the maintenance of its LGM-30 Minuteman III intercontinental ballistic missiles, a technology that could help keep the 50-year-old nuclear arsenal operational until newer, more advanced systems are available to replace it.
The U.S. Air Force is planning to use artificial intelligence algorithms to manage the maintenance of its LGM-30 Minuteman III intercontinental ballistic missiles, a technology that could help keep the 50-year-old nuclear arsenal operational until newer, more advanced systems are available to replace it.
The relevant request for information from the American military department was reported by the specialized publication Defense Express with reference to Breaking Defense.
The ground component of the US nuclear triad currently consists of 400 silo-launched missiles. They have been in service since the 1970s, with the newest one launched in 1978. With the development and introduction of the new LGM-35A Sentinel missiles fraught with difficulties, the aging Minuteman IIIs should last at least another decade.
The main maintenance problem is that technical documentation and missile health data are scattered across more than 60 different sources, from local hard drives to isolated networks. Without a single database, engineers and logisticians spend too much time searching for information and making decisions about repairs or replacements.
The task for the future AI agent will be to create a single unified interface that can effectively collect and analyze data from all disparate arrays. The department notes that previous attempts to streamline these processes using traditional software methods have failed.
At the initial stage of development, the algorithm will process only non-classified information. After successful testing and integration, the AI will receive permission to work with classified documents. Currently, the Air Force is collecting proposals from technology companies that must provide specific metrics for accelerating data processing, after which an official tender will be announced.
