FraudBrain on InstaBrain
FraudBrain is the fraud intelligence behind InstaBrain. It monitors every application written on the platform and scores its producers around the clock.
Scores advise. Every fraud review decision belongs to a named reviewer.
Served live by FraudBrain. It shows InstaBrain's monitoring status as of the time on the verification page, and pauses when monitoring stops. A monitoring status, not a guarantee.
For agents, agencies and BGAs
Here is what FraudBrain does on InstaBrain, and what it never does.
Every application written on InstaBrain is monitored and its producers are scored continuously, around the clock. The carrier's post-sale events are read too, so a pattern that only shows up after the sale is still seen.
Producers, applicants and submissions are connected into one graph, so relationships that no single application reveals come into view.
What FraudBrain flags goes to a staff review queue that refreshes every fifteen minutes. Every determination is recorded by a named reviewer, and the system checks its own outcomes afterwards.
FraudBrain watches commission advance exposure across the network, writer by writer.
How a case moves
FraudBrain scores every producer in the book around the clock and surfaces what it finds to the review queue.
A named reviewer reads the evidence. Nothing is decided by a score alone.
Every adverse action requires a verified human. FraudBrain's own lever is a reversible hold.
The seal
The seal is not a file anyone holds. FraudBrain serves it live. It shows the monitoring status for InstaBrain as of the time on the verification page, it pauses when monitoring stops (within hours, not instantly), and every seal links to that page, which says exactly what it means.
The verification page will only ever show one of three words. Not a certification. Not a guarantee.
Lit: monitoring is running for InstaBrain. Paused: it has stopped. Click the seal to read the verification page.
What we will not publish
You will not find on this page the signals FraudBrain watches, the weights it gives them, the thresholds it uses or the windows it looks across. That is on purpose. Publishing the recipe would hand it to the people it is built to catch.
What we will tell you is what happens when it works, and that it runs around the clock. Nothing about how a signal is measured ever leaves the server: carriers and organizations see outcome language only.
Questions agents ask
No. FraudBrain scores and explains. A named reviewer decides. Every adverse action requires a verified human, and the enforcement log names the person who recorded it. FraudBrain's own lever is a reversible hold.
Volume alone is not fraud. FraudBrain reads the whole network, not just how much business there is, and no adverse action happens without a verified person.
Write to [email protected] with the agent name and the dates to request a human review. If a FraudBrain hold is on the account, it can be released.
No. It means the organization showing it is monitored by FraudBrain as of the time shown. It is a monitoring status, not a certification and not a guarantee, and it says nothing about any one agent or application.
FraudBrain is built for any ledger and proven first in life insurance. Carriers and BGAs can request an architecture review at fraudbrain.ai.
Write on the platform that reads the whole network while you work the application in front of you.
Start Selling Today