What Blindsight does
Blindsight is a safety platform for AI systems, delivered as a managed cloud service. Three surfaces inspect, one turns what they found into evidence. Use whichever fit your stack.Data Security
Scan training datasets and RAG corpora for label errors, outliers,
poisoning, bias, leaked secrets, and prompt‑injection payloads. Triage,
heal, and quarantine.
Runtime Security
A drop‑in firewall for LLM traffic. Scan prompts before they reach the model
and responses before they reach the user.
DLP (endpoint)
A signed desktop agent that intercepts LLM traffic at the OS network layer
on every managed laptop. No application code changes.
Compliance
Map everything the other three surfaces observed onto SOC 2, ISO 27001,
HIPAA, GDPR, the Swiss FADP, and the EU AI Act. Then generate the report.
Pick a starting point
Quickstart
Sign in, scan your first dataset, and make your first Runtime Security call
in about ten minutes.
Data Security overview
Core concepts, severity scale, and the map of every feature.
Compliance overview
The six frameworks, what a mapping claims, and how to generate a report.
Install the DLP agent
Put the agent on one machine or a whole fleet, and know what it changes.
Runtime Security overview
Three ways to integrate the LLM‑traffic firewall, plus authentication.
DLP overview
Endpoint coverage for shadow‑IT AI traffic across the whole workforce.
How the surfaces relate
If you only need one surface today, start there. Wiring up the rest
later is additive: workspaces, Apps, API keys, observability, and audit
already know about all three.
Compliance sits on top of whichever of them you
run. It adds no scanners of its own; it maps what they already found
onto SOC 2, ISO/IEC 27001, HIPAA, GDPR, the Swiss FADP, and the EU AI
Act, and produces the document an auditor asks for.

