Use case · AI fraud investigator · $0.03 per payment assessed
AI fraud investigator
Fraud moves fast.Now your teammoves faster.
Your analysts did not train to copy numbers between screens. Toden's AI fraud investigator does the digging for them: it pulls the payment, the customer's history, the accounts they are linked to and the rule that fired, and hands your analyst one clear, sourced answer. They ask a question. They get the case. They make the call.
The workflow
The investigator that does the digging, so your analysts spend the day deciding instead of searching. This is a starting point: add, remove or reorder any step in the workflow builder.
Step 01
$0.02Transaction assessment
Scores each payment and raises an alert when a rule fires
Step 02
$0.01Account activity assessment
Scores the logins and account changes around it
Step 03
IncludedDecision rules
Your analyst decides, with the evidence already gathered
Then your rules decide
A starting rule set
- Nothing unusual about the paymentApprove
- A rule fires and the evidence needs a personReview
- The same device or address appears on a confirmed fraud caseReview
- The recipient is on your blocked listDecline
Every rule is yours to change on a visual canvas, and each result records which rule decided it. Publishing a new version needs no release.
Try the workflow builderWhat it costs
Added up from the published rate card as this page loaded. It is what one production run of the workflow above deducts from your credit.
$0.03per payment assessed
- Transaction assessment$0.02
- Account activity assessment$0.01
How it works
Every alert arrives already investigated
The moment a rule fires, the investigator opens the case and goes to work. It pulls the payment, the customer's history, their devices, addresses and payment instruments, and every account they touch, before anyone on your team has looked.
It builds the case and recommends a decision
You get a written summary, a timeline of what happened, a map of the linked accounts and a recommendation to clear, confirm or escalate, with the evidence behind each point. Ask it anything in plain language and it answers from the same record.
Your analyst decides in one click, on the record
The analyst accepts the recommendation or overrides it. The decision, the reason and the evidence it rested on go straight into the audit trail, ready for your regulator.
What powers it
Each alert comes with a plain-language summary of what happened and why it matters, with every sentence linked to the evidence it came from.
Integrate
Mount the SDK with the workflow you published, or send a hosted link and write no code. The result arrives by signed webhook, and changing the workflow never needs a release.
import { TodenVerification } from "@todenhq/sdk-react"; export function Verify({ userId }) { return ( <TodenVerification apiKey="pk_test_..." workflowId="wf_AbC123dEf456" userId={userId} onSubmit={(s) => save(s.verificationId)} /> );}{ "event": "verification.completed", "data": { "verificationId": "ver_8fk2...", "status": "verified", "assuranceLevel": "gov_db", "idType": "bvn", "country": "NG", "dataMatch": true, "facialMatch": true, "facialConfidence": 91 }}Products involved
This solution combines products you can use on their own too.
Related use cases
Coverage
Pick a country to see the IDs you can verify there, which documents carry a chip we read, and whether business checks are available.
Frequently asked
It recommends, and your analyst decides. If you choose, it can close clear false positives on its own within limits you set, and every one of those is logged with its reasoning for review.
AI fraud investigator,in one workflow.
The sandbox opens when you sign up. Usage-based pricing, no sales call.