美国国防体系正在扩大人工智能在军事决策中的应用，尤其通过Palantir平台整合聊天模型。自2017年起，Palantir成为五角大楼“Project Maven”主要承包商，该项目利用卫星图像与计算机视觉识别潜在敌方系统，并由国家地理空间情报局管理。Maven已在陆军、空军、太空军、海军、海军陆战队及美国中央司令部等多个部门部署，用于分析情报、识别目标并支持军事行动规划。

Palantir的软件演示显示，AI聊天助手可在军事平台中辅助分析情报并生成作战选项。例如系统可通过雷达图像检测“异常敌方活动”，随后帮助分析人员判断敌军单位类型，并生成多种作战方案。一个示例展示AI在数秒内提出三种攻击方案：空中打击、远程炮兵或战术小队行动。相关系统还能生成部队行军路线、分配电子干扰设备并汇总战场信息，从而加快战术决策流程。

AI还被用于情报分析和报告生成。示例显示，Claude模型可自动生成约200字的军事影响摘要并创建互动数据面板，整合公开信息或政府内部数据库。过去类似分析通常需要研究人员花费约5小时收集资料并撰写报告。通过自动化数据整合与分析，AI系统可显著缩短情报处理时间，但同时引发关于军事自动化、监督和技术依赖的政策争议。

The US defense system is expanding the use of artificial intelligence in military decision processes, particularly through integration within Palantir software platforms. Since 2017, Palantir has served as the primary contractor for the Pentagon’s Project Maven, which uses satellite imagery and computer vision to identify potential enemy systems and is managed by the National Geospatial Intelligence Agency. Maven has been deployed across multiple branches including the Army, Air Force, Space Force, Navy, Marine Corps, and US Central Command to analyze intelligence, detect targets, and support operational planning.

Software demonstrations indicate that AI chatbot assistants can help analysts interpret intelligence and generate operational options. In one scenario, radar imagery processed by AI detects unusual enemy activity, after which the chatbot assists analysts in identifying the likely enemy unit and proposing actions. The system can produce three attack strategies within seconds, such as deploying air assets, long-range artillery, or tactical teams. It can also generate troop routes, assign electronic jamming equipment, and compile battlefield information, accelerating tactical decision cycles.

AI systems are also used to produce intelligence assessments and analytical reports. Demonstrations show the Claude model generating a roughly 200-word summary of a military operation and building interactive data dashboards using public or internal datasets. Comparable analysis previously required analysts to spend about five hours collecting sources and writing reports. By automating data integration and interpretation, these tools significantly reduce intelligence processing time while raising policy debates about military automation, oversight, and technological dependence.