Space Threat Intelligence Synthesis Engine

Problem

Today’s space and missile threat information is spread across a variety of sources. Existing tools struggle to distinguish closely spaced objects, track emerging threats, and keep threat models current. A more effective approach is needed to give operators and automated systems a clear picture. The resulting insights must be available quickly enough to inform time-sensitive operational decisions.

The Department of War is looking for AI-based software tools to turn fragmented sensor data into a continuously updated understanding of space and missile threats fast enough to drive operational decisions.

Desired Solution Attributes

The desired solution will ingest and fuse massive streams of multi-source data, including live video, satellite imagery, radar and sensor feeds, geospatial data, and classified intelligence reports. By automating advanced cross-data correlation, the architecture aims to uncover hidden patterns, complex operational relationships, and predictive threat behaviors far beyond the capabilities of human analysis alone. Designed around open standards, the solution will continuously update weapons-system performance models in real time. It will deliver dual outputs: intuitive, human-readable visualizations for frontline operators and low-latency, machine-to-machine APIs to drive automated command-and-control (C2) workflows.

Compelling solutions must have:   

AI/ML Core Capabilities

• Leverage AI/ML, computer vision, predictive analytics, semantic technologies, or knowledge graphs to combine Multi-INT feeds into unified situational awareness.
• Ensure outputs feature transparent reasoning, confidence scoring, and data provenance.
• Ensure capture of decision-support analytics into workflows and training models, output in non-proprietary formats.
• Provide documentation for Government evaluation of algorithms, models, training/test data, and generation procedures.

Threat Characterization & Attribution

• Improve detection, classification, and attribution timelines, assigning events to specific threat actors, platforms, or patterns.
• Provide high-confidence notifications among closely spaced or co-located objects in degraded data environments.

System Speed, Latency Considerations

• Ingest, enrich, and correlate structured, unstructured, streaming, geospatial, imagery, video, and intelligence data streams.
• Operate across high-speed/low-volume and low-speed/high-volume threat scenarios with real-time, low-latency processing.
• Latency (Data Time of Arrival to Display/Query): Threshold of 5 seconds with an objective of 2 seconds.
• Throughput: Nominal rates of ~20–30 MB/min with burst handling up to ~5 GB/min.
• Store and retrieve multi-year data to support historical, real-time, and post-event analysis.

Security, Infrastructure & Enterprise Interoperability

• Deliver scalable DevSecOps, classified data protection, CUI management (performance and tagging), and access controls supporting multi-partner operations.
• Deploy containerized, cloud-native architectures that permit independent component upgrades.
• Comply with NIST SP 800-171, integrate PKI certificate-based authentication where needed, and enforce zero Critical/High CVEs (zero CVEs preferred).
• Maintain interoperability with government data pipelines, C2 architectures, mission systems, and visualization tools through modular reuse.

Future Sustainment

• Generate actionable, confidence-scored alerts supporting analyst-in-the-loop, human-on-the-loop, or automated workflows.
• Offer a clear path from near-term prototype demonstration to enduring mission workflows with value-added capabilities.

Solution Differentiators include:

AI/ML Core Capabilities

• Capability to generate or use synthetic data, modeling, simulation, or digital representations to evaluate current and future threat scenarios.
• Demonstrated ability to evaluate and validate model performance against representative operational test cases, algorithms, and benchmark data libraries.

Threat Characterization & Attribution

• Demonstrated track record or benchmark metrics showing significant speed and accuracy improvements in event detection, characterization, and attribution compared to existing baseline systems.

System Speed, Latency Considerations

• Low-latency processing suitable for real-time or near-real-time situational awareness.
• Ability to store and retrieve multi-year data for pre, current, and post analysis. 
• Proven empirical performance maintaining sub-2-second latency and system stability under peak burst data conditions (~5 GB/min).

Security, Infrastructure & Enterprise Interoperability

• Advanced MLOps and continuous model lifecycle management supporting automated deployment, monitoring, retraining, and sustainment in operational environments.

Future Sustainment

• Mission-Critical Support: Mature sustainment model capable of maintaining 24/7 availability, rapid incident resolution, and resilient operations during degraded conditions.

Vendors are welcome to apply individually or in partnership. The Government may also request teaming arrangements amongst solution providers. Vendors are expected to demonstrate their solution in an unclassified test environment as part of the Commercial Solutions Opening.

Vendor Solution Brief Submission Options 

Vendors have flexibility in how they submit their solution briefs, which can be proposed either independently and/or through a teaming arrangement. Vendors who choose to provide a teaming arrangement must identify and should elaborate on the role each partner will play in the solution.    

Process 

Submissions will be evaluated in accordance with CSO HQ084520SC001 available on https://DIU.mil and https://SAM.gov. Vendors selected for Phase 2 will receive an amplification letter with expanded details to help inform their Phase 2 pitches. Those vendors will be expected to provide their Rough Order of Magnitude (ROM) cost breakdown. The vendorsʼ Phase 2 pitch shall address the Phase 2 evaluation factors contained in CSO HQ084520S C001. DoW requires companies without a CAGE code to register in SAM https://SAM.gov if selected for an agreement award. The Government recommends that prospective companies begin this process as early as possible.