Pattern Intelligence Engine
UAS intelligence with the evidence, coverage, and uncertainty attached.
Patterns separates current assessment from historical record. Scores prioritize research; linked sources, provenance, confidence, and limitations explain what the system actually knows.
Public data status
Loading
— fresh required-current datasets
— required-current datasets
— cataloged public datasets
— oldest current coverage date
Loading the public data catalog…
Current assessment
Freshness-gated views of what changed, what deserves review, and why the signal moved.
PIR
Priority intelligence view Local Rank the current public corpus by declared FPV, compliance, procurement, adversary, C-UAS, DFR, or autonomy terms without hiding the full datasets. What matters to my role?
›
01
Daily intelligence brief Synthesis The most important changes, evidence, caveats, and watch items in one reading path. What changed today?
›
02
Signal board Current Procurement, regulatory, supply-chain, and gray-zone signals ranked for review. What requires attention?
›
03
Threat actor signals Triage Actor-centered article mentions, source diversity, TTP context, evidence, and attribution limits. Who appears in the reporting?
›
04
TTP and counter-signal gaps Indexed Observed UAS techniques compared with indexed defensive-procurement evidence. Where is indexed defensive signal weak?
›
Evidence and relationships
Underlying records, technical observations, and graph relationships that support or challenge an assessment.
BOM
Adversary teardown BOM Publicly documented components, countries, and platforms found in captured UAS samples. What was physically observed?
›
MIR
Component mirroring Component-type similarity used as a research lead, never as proof of transfer or misconduct. Which technical profiles resemble each other?
›
NET
Sanctions-evasion signals Public relationship paths with explicit relationship strength and allegation caveats. Which relationships deserve validation?
›
ASK
Ask PIE Cited retrieval Turn a question into a deterministic evidence packet with source links, dates, match reasons, coverage, and limitations—without a generated conclusion. What does the indexed evidence show?
›
SRC
Raw corpus search Search indexed flags, actors, entities, predictions, and public reporting without an evidence-packet interpretation layer. Show every literal match.
›
History and accountability
Historical snapshots remain useful only when the start date, gaps, and grading rules are visible.
Reading discipline
How to read Patterns
The system is strongest when assessment, evidence, and coverage are read together. A score alone is a prioritization aid, not a conclusion.
1 Read the assessment What changed, over what window, and which dimension moved?
2 Open the evidence Review independent sources, direct records, contradictions, and provenance.
3 Check coverage and confidence Confirm freshness, collection gaps, source concentration, and attribution limits.
Important: Article mentions are not confirmed incidents. Relationship proximity is not an allegation. Weak indexed procurement evidence does not prove that no program, funding, or capability exists.
Analytic Signal — Not an Allegation
AI-assisted public-source analysis. Not legal, procurement, export-control, airworthiness, or operational advice. Compliance labels and risk scores are informational and may be incomplete or stale. Verify against official sources, vendor attestations, and qualified counsel before relying on them.
Risk labels are analytic signals, not allegations of unlawful conduct. They reflect public-source indicators, confidence levels, and available data at the time generated.