We turn public data into public safety — finding missing persons and exposing environmental injustice with detection systems that are precise, private, and free to the communities they serve.
Each one connects public databases that already exist — but that no one else joins together — and gives the answer away.
Distributed missing-persons detection. Submit a photo; Guardian compares it against NamUs, FBI, and Interpol records using facial-vector matching. Images are deleted the instant the comparison finishes — zero biometric retention.
Environmental-health risk lookup. Enter an address, ZIP, or census tract; Sterling cross-references EPA contamination, Census demographics, and Medicaid cost-shift to show whether a neighborhood is paying the health price of nearby pollution.
The machine-learning workshop behind the systems. Geppetto carves the agents — and the myths that make them make sense. Every detector begins as a story about how the world holds together, then becomes a gate that only speaks when the evidence forces it to.
A domain-agnostic template of how order shows up — the story that tells the machine what "coherence" looks like before it ever sees data.
The myth becomes a forcing gate: it commits only when public, mechanical evidence compels it, and abstains — honestly — when it doesn't.
The agent ships as a public-good system — Guardian, Sterling, and the next ones — anchored, auditable, and free to the people it serves.
Statistics refresh from the live API. See API status.
The information needed to find a missing person or prove environmental injustice is sitting in public databases right now — unconnected. Black Swan Technologies builds the connecting infrastructure and gives the results away. The core technology is held in the public trust, funded by grants, donations, and license fees. This work is not abstract for us; it is a debt we are paying.
Our engine stacks weak correlating factors across many data layers to reveal patterns invisible to linear analysis — then refuses to overclaim. A correct "no" is a credential.
Our work runs on federal grants, philanthropic donations, and license fees. Every dollar moves the technology closer to the people who need it.
Get Involved