Risk, KYC and compliance

Scores you can explain, cases you can close.

Configurable rules produce a 0–100 score and a risk level for every player, recalculated nightly and on events. Analysts see the score move, open a case, assign it, escalate it and resolve it against an SLA. Auto-actions ship report-only until you switch them on.

Launch your brand Book a demo

Risk score 0–100Duplicate accountsin-house fraud scoringKYCGeo and GPSResponsible gaming

console.gamebridge.io/admin/risk/cases

GB Console

Players

Dashboard
Players
Payout requests

Growth

CRM campaigns
Bonus Engine
VIP management
Segmentation
Packages
Banners

Control

Risk cases
Reports
Analytics
Activity log

Casino

Games
Originals
Slot Forge
Poker

Working brand

Nova ▾

Risk management / Cases

⌘K

Risk · Cases

Mode: report-only

Player pl_8c1e · risk score

Same device+32

Payout velocity+24

Geo mismatch+12

Chargeback hist.+10

HighDeposits blockedKYC: verified

Case	Trigger	Priority	Status	SLA
#4821	Duplicate device · 3 accounts	High	Investigating	2h left
#4819	Payout velocity · 4 in 24h	Medium	Open	6h left
#4815	Fraud score 91 at login	High	Escalated	Breached
#4812	Bonus abuse pattern	Low	Resolved	—

Duplicate signals: email · phone · device · address · name · IP · location

Verified in the console

0–100
risk score from configurable rules, nightly and on events: Report-only
auto-actions ship off until you switch them on: Device · ID · location
signals behind duplicate-account detection: Graded
suspension reasons, timed blocks and surgical action blocks: Safe by default
payments stop when a signature or secret is missing

What the system covers

Identity, behavior, geography and the player's own limits.

Risk is a workflow here, not a verdict. Signals feed a score, the score feeds a case, and the case ends in a documented, logged action.

Risk scoring and cases

Score 0–100 from configurable rules, mapped to risk levels
Nightly recalculation plus event-driven updates when something changes
Whitelist and manual override, both audited
Auto-actions are report-only until enabled, so the team sees what would have happened first
Cases: open → investigate → resolve, with assignment, escalation, SLA and notes
Risky Players, Duplicate Accounts and in-house fraud scoring Lookup screens in the Risk group of the console

Identity and device

Duplicate Accounts across email, phone, device, address, name, IP and location
in-house fraud scoring at registration and login, with device fingerprinting through the native plugins and block-on-decline
ID document and selfie with liveness; callbacks hardened
MaxMind GeoIP as a second opinion on location
Admin 2FA with an authenticator app, backup codes and lockout

Geo and responsible gaming

Country and state allow/deny lists, a restricted-region page and GPS re-check for state-level rules
Deposit, loss, wager, single-bet and session limits with per-brand defaults
Reality checks, cooling off and self-exclusion from one day to permanent
Consent matrix: Email, SMS, Push, Phone, Post and Partner × Casino and Sports, with one-click unsubscribe
Suppressions for bounces, complaints, SMS STOP, invalid and manual, global or per brand, with audited lift
GDPR purge with typed confirmation; every admin action in the Activity Log

Bonus abuse controls

Five limits enforced on every grant path: per player, per day, global, budget and cooldown
Velocity checks at bonus time
Duplicate Accounts review before a redeem is approved
in-house fraud scoring block-on-decline at registration keeps declined signups out of the player base
Referral program with eligibility rules and fraud blocking

Sweepstakes compliance

The rulebook is a brand setting.

Business model is chosen per brand. Sweepstakes gates AMOE and the Store-plus-Redeem cashier mapping; Real Money and Crypto switch to Deposit and Withdraw. The same console runs both.

AMOE mail-in entries with 8-character barcode postcards and an admin queue for Requests and Settings
Gold Coins (GC) for play and Sweeps Coins (SC) for prizes, redeemable after playthrough and KYC, on one ledger
Sweepstakes rules pages served per brand
Country and state allow/deny with a restricted-region page

State-level geo with GPS re-check where a country lookup is not enough.

How a case moves

Signal, score, case, resolution.

Step 1

Signal

Registration and login pass in-house fraud scoring; ID and selfie are verified with liveness; MaxMind and GPS confirm location; duplicate-account matching runs across device, identity and location signals.
Step 2

Score

Configurable rules produce the 0–100 score and a level. Nightly recalculation catches slow drift; event-driven updates catch sudden change.
Step 3

Case

An analyst opens or is assigned a case, investigates with notes, escalates if needed and works against an SLA. Report-only auto-actions show what an enabled rule would have done.
Step 4

Resolution

Resolve with a documented outcome: whitelist, manual override, a graded suspension reason, a timed block or a surgical action block. All of it lands in the Activity Log.

Teams

Compliance, fraud and support work the same case.

Evidence you can hand to an auditor.

Activity Log for every admin action
Consent matrix and suppressions enforced on every send
Responsible-gaming limits, reality checks, cooling off and self-exclusion
Certification-ready RNG dossier for GLI-11/19, BMM and iTech Labs submissions

Evaluator questions

What compliance teams ask first.

Does the platform block fraud automatically?

It scores and routes. Auto-actions exist but ship report-only until you enable them, so analysts see what a rule would have done before it does it. In-house fraud scoring with block-on-decline at registration and login is the hard gate at the door.

Are you GLI-certified?

Not yet. The RNG dossier and submission handbook are prepared for GLI-11 and GLI-19, BMM and iTech Labs. We say certification-ready and mean exactly that: the paperwork and the reproducibility artifacts exist; the certificate does not until a lab issues it.

Can we detect one person running several accounts?

Duplicate Accounts matches on email, phone, device, address, name, IP and location and lists linked profiles for review. In-house fraud scoring adds device fingerprinting at registration and login. Bonus-time checks are velocity-based; we do not claim automated multi-account bonus-fraud detection.

Which KYC vendor is integrated?

ID verification, with ID document and selfie, liveness and hardened callbacks. Other vendors are on request; see Integrations for what is live and what is not.

How is responsible gaming handled for real-money brands?

Players set deposit, loss, wager, single-bet and session limits, get reality checks, and can take cooling off or self-exclude from one day to permanent. Defaults are per brand, and the console shows the same controls on the player's Responsible Gaming tab.

How do you handle state-by-state rules in the US?

Country and state allow/deny lists drive a restricted-region page, MaxMind provides a second opinion on IP, and a GPS re-check runs where state-level precision matters.

Scores you can explain, cases you can close.

Identity, behavior, geography and the player's own limits.

Risk scoring and cases

Identity and device

Geo and responsible gaming

Bonus abuse controls

The rulebook is a brand setting.

Signal, score, case, resolution.

Compliance, fraud and support work the same case.

Evidence you can hand to an auditor.

Signals in one place, decisions in one workflow.

Act on a player without breaking their account.

What compliance teams ask first.

Systems on either side of a case.

Payments and ledger

Operator console

Markets

Lancez une marque qui tourne comme un logiciel et règle comme une banque.