Security investigations require more than an AI prediction.
VIGIL is being built around evidence, reviewability, and uncertainty.
Real-world footage is rarely clean.
All of these can affect what a system can reliably determine. VIGIL is designed to distinguish confirmed evidence from uncertain continuity — rather than smoothing over the difference.
When available evidence cannot reliably establish continuity, VIGIL can preserve the result as unresolved rather than silently forcing an identity decision. That matters especially when a person:
DISAPPEARS BEHIND ANOTHER SUBJECT
LEAVES THE CAMERA'S FIELD OF VIEW
APPEARS LATER, POSSIBLY ELSEWHERE
CANNOT BE OBSERVED CLEARLY
VIGIL is an investigation tool, not a verdict machine
AI can help surface, organize, and connect evidence, but consequential conclusions should remain reviewable by authorized personnel.
Traceable evidence
Every result connects back to a video source, camera, timestamp, observation, relevant clip, and timeline position — so investigators can inspect the footage behind a result rather than relying solely on an AI-generated conclusion.
No mysterious score
The goal isn't a confidence percentage floating with no explanation. It's helping an investigator see exactly what the footage does and doesn't support.
Reviewable by design
Uncertain identity and continuity decisions can remain open for review rather than being silently forced into a match — the investigator makes the final call.
Camera access is sensitive infrastructure
Only authorized users should be able to access cameras, investigations, and evidence.
VIGIL detects the signal. Your team makes the decision.
VIGIL should not automatically decide that someone is dangerous, suspicious, or committing a crime. It surfaces observable events, grounded in evidence, for a trained operator to verify.
More alerts don't necessarily mean better security. Alert quality is measured through operator feedback — confirmed incident, relevant, not relevant, false alert — without silently changing safety-critical behavior behind the scenes.
We won't claim controls we haven't verified
VIGIL's deployment and security architecture will depend on the production environment and customer requirements. We are developing the platform with enterprise security requirements in mind.
Detailed information regarding deployment, access controls, data handling, retention, and security practices will be provided as those production controls are finalized and verified. We will not claim certifications or security controls that have not been independently established.
You stay in control of your cameras and your data
Organizations using video intelligence remain responsible for ensuring their camera systems and use of video comply with applicable laws, policies, contractual obligations, and privacy requirements.
VIGIL is designed as a tool for authorized security and incident investigation — not as a substitute for human judgment.
Talk to us before you commit to anything.
We'd rather walk you through exactly what's built, what's in progress, and what isn't in place yet.