Geospatial intelligence artificial intelligence (GEOINT AI), refers to the integration of artificial intelligence (AI) into the field of geospatial intelligence (GEOINT). Geospatial data includes information that has a geographic component, such as maps, satellite imagery, GPS data, and any data that includes location-based elements. GEOINT involves the collection, analysis, and interpretation of geospatial data to inform decision-making in areas such as national security, defense, disaster response, and other critical sectors. When AI is applied to GEOINT, it enhances the ability to process and analyze large volumes of geospatial data more efficiently and effectively.

NGA employs these advanced analytics to decrease the time needed to sift through the staggering amount of data to discover and provide tip-off and alerts to systems, collectors and analysts. This allows analysts to interact effectively with the data, discover new objects and ask intelligence questions that provide the context necessary to deliver decision-ready expert GEOINT.  

In other words, GEOINT AI provides trusted GEOINT at speed and scale. NGA leads the GEOINT Enterprise in exploring and applying all forms of AI, with an emphasis on the fields of Machine Learning (ML) and Computer Vision (CV).
 

AI is a revolution in warfare and intelligence creation. The pace of available information is not slowing down. Over the next 5-10 years we will experience a potential tripling of GEOINT data when considering all of the government and commercial satellites programs that have or will achieve full operational capability. With the rapid increase of collection assets, how do we manage this deluge of data to gain a competitive edge?

Machines make it possible to liberate analysts to focus on critical projects and intelligence assessments.

NGA aims to bring together disparate, often siloed communities of AI experts, data and computer scientists, as well as tradecraft professionals, to improve AI model performance, develop standards, and lead interoperability efforts for the GEOINT community. 
 

NGA will lead the GEOINT Enterprise in applying AI to the GEOINT mission, enabling decision advantage for the warfighter, policymakers and mission partners. NGA's primary goal is to produce increasingly accurate GEOINT data and AI models that will generate high-quality and useful detections and reports to be applied to defense and intelligence missions. 

NGA must give our mission partners the information they need, when they need it. Operating near the end user facilitates a better understanding of requirements, and enables the ability to quickly develop and deploy solutions. Continued partnerships enable the fusion of intelligence and operations information into workflows that support the warfighter. 

1. Utilize high quality CV that improves positive ID, geolocation, and speed.  
NGA will improve and scale existing overhead imagery Broad Area Search-Targeting and Full Motion Video Lines of Effort by employing novel algorithms and techniques in order to deliver target detections from new modalities, deliver CV models to military service partners for use on autonomous systems, and automate geolocation accuracy of detections.

2. Integrate AI into the analytical workforce.  
NGA is striving to create common tooling, techniques, and standards by improving Structured Observation Management conversion to labels, increasing analyst feedback to models, and exploring synthetic learning and labeling.  AI reasoning and search features will be enabled by recent advancements in generative AI models, and will expand the discovery of new types of objects.

3. Assimilate AI into informed collection orchestration.
NGA currently tasks overhead GEOINT collection 24/7, with very little fanfare. An established service will continuously present options for the collection managers, based on constellation constraints, standing needs, and dynamic events.

4. Implement an enterprise AI infrastructure.
NGA will optimize imagery services, data storage, and data access, as well as compute to lower costs and increased speed. A common platform will incorporate data management for labels, models, and detections for our analysts and our customers.  

MAVEN 

NGA Maven was established in 2017 as the Pentagon’s flagship AI project to integrate AI into military workflows. The GEOINT aspects of Maven were entrusted to NGA in January 2023. Its state-of-the-art computer vision and AI capabilities are now integrated into various military analytic workflows to automatically detect, identify, characterize, extract, and attribute features and objects in imagery and video. Being able to sift through the barrage of data, and discern an object from a non-object – with high accuracy, based on unique behavior, at the speed of conflict – is key to maintaining our decision advantage. Ultimately, Maven provides trusted GEOINT at speed and scale for object recognition. 

NGA Maven is already producing large volumes of computer vision detections for warfighter requirements across multiple operational locations. It’s generated millions of data labels and lowered latency detections. Maven data and detections are fed into multiple other platforms and stands as an important thread in a tapestry of connected sensors from all branches of the armed forces in a unified AI network. As the lead for the GEOINT Enterprise, NGA takes the responsibility seriously to make our information as accessible as possible, especially to our operators – and NGA Maven is leading the way in data accessibility.

NGA Maven has also already successfully aided our Combatant Command partners with several key objectives, including:

• Increasing the speed to locate objects and directing analysts to abnormal or significant activity in near real-time. 

• Providing the ability to scale to locate more objects and analyze more images. 

• Improving the accuracy of our models to meet or exceed human detection, classification, and tracking performance.

• Enhancing the interoperability to interface and share data, models, and detections with partners for use in their existing tools.

ASPEN

ASPEN stands for Analytic Services Production Environment for the NSG (National System for GEOINT). It is a program of record, with a five-year modernization effort for intelligence production that began in May 2023.  It is meant to address the near tripling of GEOINT data coming to NGA’s analysts.

ASPEN provides trusted GEOINT at speed and scale for warning, and the goal is to provide that speed without compromising accuracy or precision. ASPEN’s premise is to let the machine do what the machine does best, while having the analysts do what analysts do best. This involves enhancing, integrating and scaling analytic capabilities to automate and modernize GEOINT analysis.   

ASPEN will result in deeper analysis, enabling a data-centric analysis of monitoring known behaviors at known locations and searching for unknown behaviors at unknown locations. This delivers pattern of life analysis that baselines adversary tactics, techniques and procedures, characterizes activity and anticipates events to get ‘left of launch' and provide advanced warning for NGA’s partners, customers, and stakeholders. 

NGA is at the forefront of ensuring the ethical and responsible use of AI. We’re establishing a certification program that we call the GEOINT Responsible AI Training program – GREAT for short.  It will teach developers and users to assess risk and responsibility in applying AI to GEOINT.  It also will be consistent with national guidance, informed by leading AI scholarship, and ensure that our values, collateral damage mitigation, and the Law of Armed Conflict are baked into our use of AI.