One explainable AI risk engine that flags fraud, scams and insider loss at the decision point, so your team can stop them before the money moves.
A CEO-impersonation email asks for an urgent, confidential transfer. Vilkax screens the sender, the domain and the intent, and flags it before the money moves.
Vilkax Protect
demo
CEO · Urgent payment
ceo@vilkax-corp.co: "Wire €48,000 to a new supplier today, keep it confidential."
BEC · CEO-impersonation
Look-alike domain (vilkax-corp.co, not vilkax.com) plus an urgency and secrecy pattern.
Grouped the way risk, compliance and security teams think. Every match returns one explained verdict: pattern, evidence, confidence.
The money's gone before the alert fires. We catch it at the yes / no / review moment.
Caught by B2B fraud ML + the Protect APIStarter+
Structuring, shells and sanctioned counterparties hide between systems. We surface them as decision-support, including scam proceeds a single-lane vendor would never see.
Caught by B2B fraud ML + anti-scam intel + the OSINT moatPro+
One sanctioned hop can poison a payout. KYT screens any wallet or transaction in real time and returns an allow / review / block verdict with the evidence behind it.
Caught by KYT (Know-Your-Transaction) + the investigation enginePro+
Your hiring funnel, your approvals and your endpoints are the newest attack surface. We read the pattern and support your security and audit teams. Assessment only, never surveillance of a person.
Caught by B2B fraud ML + anti-scam intel + the AI Security ShieldPro+
Most loss starts with a message and a look-alike domain. We catch the lure before it lands a wire or a credential.
Caught by OSINT moat + the domain pulseStarter+
A device, wallet or payee reused across supposedly independent accounts is a fraud ring. A supplier shared by half your vendors is a single point of failure. We read both on the exposure graph.
Caught by entity intelligence + the exposure graphPro+
Assessment, not accusation. Vilkax detects patterns, never labels a person a criminal, and never auto-acts. High-stakes calls route to human review with the evidence. Every clear verdict carries a false-clear risk: thin coverage is flagged, not silenced.
Spans human → financial → on-chain in one model. Because Vilkax runs the consumer scam moat, we see a pig-butchering victim's payout land as scam-proceeds in a customer's AML picture - the consumer scam and the financial exposure together, an angle no single-lane vendor has.
Every verdict ships with reasons, evidence and a confidence score. Your team and your regulator both see exactly why.
Catches threats from day one with built-in intelligence. No model to train, no history needed.
When a source isn't live, the verdict says so. It never returns a clean result from missing data.
Stop invoice and CEO fraud before money moves: a payee-change hold with an out-of-band checklist, where whoever raised a hold cannot be the one who verifies it; a phishing report button that tells the reporter the verdict; email domain checks (SPF, DMARC, MTA-STS); and a monthly report you can forward to the owner.
Vilkax is SOC-lite, not a SOC platform. It adds scam and fraud signals, staff phishing triage, impossible-travel and connected-app findings, response playbooks, SCIM provisioning, OIDC sign-in (SAML is not available yet) and a hash-chained audit export for your SIEM. If you already run a full SOC platform such as ClearSkies, keep it: Vilkax sits alongside through the audit export and signed webhooks.
A scam-intelligence API with idempotent screening; sandbox keys that return fixed test answers and write nothing; signed webhooks (HMAC with a timestamp) with a delivery log and replay; and Protected Customer Link, in pilot, so a customer can check on their own device whether a call that claims to be you is real.
Family and elder protection modes, Kin circles, a carer rule that asks a chosen person before a large payment, and a panic alarm whose emergency relay is free on every plan. There is no institutional plan yet: tell us what you need and we will say honestly what fits.
Both reach the team and we reply by email.
Assessment, not accusation. Auditable, PII-safe, built in-house.
No password. We email a secure sign-in link straight to your control panel. Up and running in under a minute.
Already with us? Open your control panel.
Tell us what you'd like to see and someone from the team will reply to set up a walkthrough.
Tweaks