Unstructured wins Army SBIR award for anomaly detection

On September 22, following selection through the Army’s xTechIgnite competition, Unstructured announced that it has been awarded a $2 million Army Small Business Innovation Research (SBIR) Direct-to-Phase II contract under topic A254-018, Novel AI Techniques for Insights in Various Environments (NATIVE). During this 18-month effort, Unstructured will support Product Manager (PdM) Artificial Intelligence Operations and Services (AIOPS) under the Capability Program Executive Command and Control Information Network (CPE C2IN), operationalizing AI-ready multimodal data pipelines that ingest and preprocess sensor data to detect anomalies and surface unknown patterns. The objective is concrete: deliver AI capabilities that compress the sensor-to-decision timeline, reduce cognitive load on Soldiers, and sharpen situational awareness.
This effort centers on solving a central problem in modern command and control: transforming voluminous, unstructured multimodal sensor data into AI-ready data that aids Commander’s decision making. That data is delivered in non-proprietary JSON schema, ready for real-time consumption by downstream models and agents. This effort extends Unstructured’s agent-ready data pipelines beyond documents into sensor modalities that define multi-domain warfare. In parallel, Unstructured will develop critical anomaly detection capabilities across heterogenous sensor data. These solutions will have direct applicability across warfighting functions by surfacing critical information.
This award reinforces Unstructured’s commitment to solving the data readiness challenge for GenAI in mission environments, so that Soldiers have reliable GenAI tools and agents in the moments that matter, the company said.
Source: Unstructured
IC News delivers the situational awareness you need to get ahead and stay ahead in the IC contracting space. Subscribe today for full access to 10,000+ articles, plus new articles each weekday.







