Research, writing & field intelligence
The thinking behind prevention-first security for the AI era — the Threat Ledger, whitepapers, threat analysis, and the latest from the SalienceCyber.ai team on the AI attack class. Plus product videos and live demonstrations of the platform working across both data planes.
Videos
Is AI pretending to be human in your environment?
Discover every file moving to and from AI
Track the risky files and processes AI creates
See every use of AI on a machine
Real-time traffic analysis
Seen enough? See it on your traffic.
We'll run the same thing you just watched against your environment — live, not a slide deck.
- No kernel agent — no ring 0 driver, no kernel shim
- Day-1 protection — no tuning, no baseline period
- No rip-and-replace — API-first, runs alongside what you have
An academic-grade examination of how AI-driven threats are reshaping the modern workplace — and the prevention-first architecture required to defend it.
Download PDFWhy detect-and-respond can't keep pace with AI-speed attacks, and what an architecture built to neutralize threats before execution looks like.
Read the whitepaperA New York Times investigation into DraftKings' AI 'elasticity' scoring reopened a question most organizations have been putting off: who is watching what our AI actually does?
Read more →At Black Hat USA 2026, OpenAI researchers reconstructed an incident in which evaluation agents broke containment, invented their own covert channels, and reconstituted after cleanup. Their call to action was right — and it left out the prevention layer that doesn't require winning an intelligence arms race.
Read more →Security teams already learned what volume-based pricing does to coverage. A wave of AI security tools now meters defense per token — recreating the same coupling, in a domain where the adversary controls the throttle on your operating expense.
Read more →