Skip to content
Try one-tap loginBook a demo

Verify customers, sign them in, stop fraud.

One SDK gives banks and fintechs one decision, with its reasons, at every sign-up, login and payment.

Book a demoTry one-tap login

Illustrative: the customer signs in on her phone with nothing to type. The DeepID console logs the phone as recognised, her number confirmed by silent network verification, the SIM and phone bound to her and the session read, then allows the login.

Every check, as it runs. On one customer’s phone.

Follow Ananya from sign-up to payment through DeepID’s four families of modules. Choose any check to see what DeepID reads and what it decides.

The 15 modules, in four families

  • SNVConfirmed by the network
  • Code sentNone
  • DeviceSame as at sign-up

AuthenticationSilent network verificationillustrative session

Authentication

Sign in with one tap, on her own phone.

Silent network verification (SNV) confirms the number in the background, and device binding keeps the account on her own SIM and phone.

  • SNVConfirmed by the network
  • Code sentNone
  • DeviceSame as at sign-up

Silent network verification. One-tap login: the mobile network confirms the number silently. No code to wait for, type or steal.

How one-tap login works

SIM and device binding. Device binding ties the account to her own SIM and phone, so payments work only from them.

How binding works

Re-binding a new phone. When her SIM moves to a new phone, her face is checked first, then the new phone is bound.

How re-binding works

Verification

Know who is opening the account.

Aadhaar, documents and a live face, verified at onboarding and again whenever it matters.

  • AadhaarVerified with UIDAI
  • PANName matches
  • Sanctions and PEPClear

Identity and eKYC. Aadhaar, DigiLocker and PAN, verified at the source.

How eKYC works

Liveness and face match. A live person on a real camera, not an injected stream or a deepfake.

How liveness works

Also documents and IDs, video KYC, business verification, AML screening and bank account verification.

Intelligence

Know the phone, her rhythm and the payee.

Device, behaviour, scam and mule signals, read in every session. Every check can pass while she is coached on a call, so DeepID reads the session and the payee too.

  • Device IDDV-8F2K-41
  • First seen412 days ago
  • AppGenuine, not rooted

Device fingerprint. The same device ID after a reinstall or a factory reset, and the tampering around it: emulators, rooted phones, hooking tools, repackaged apps.

How device intelligence works

Remote access. Screen sharing and remote-control apps, while money moves.

How remote access is caught

Her own rhythm. Typing and touch, matched to the way she uses the app, learned from her first sessions.

How behaviour is read

Pasted and dictated input. A payee pasted in and long pauses, as if someone is telling her what to type.

How coaching is caught

An active call. A long call with an unknown number, during a payment.

How scam signals work

Mule payees. A payee linked to many unrelated senders, cooled off before it is paid in full.

How the payee graph works

Decision

One decision, with its reasons.

Allow, warn, step up, cool off, hold or block, by the policy your fraud team writes.

  • ActionHold, and call her back
  • ReasonsActive call, shared screen, mule-linked payee
  • Sent toHer app, your FRM, the case queue

One decision. One of six actions, with the reason codes behind it: what she reads on her phone, and what your console shows.

How the decision works

Into your FRM. The same record in your app, your FRM and your case manager, with the same reason codes.

How it reaches your FRM

Journeys your team changes. Onboarding, login and payment journeys, composed from the modules and changed by your fraud team without a release.

How the journey builder works

A takeover steals the session. A scam steers the customer.

Stronger login stops the first. Only reading the session stops the second, because the customer passes every check herself.

How DeepID reads each moment

Digital arrest

Scams

Hold, then call back

02Scams

Hold the coerced payment before it leaves the account

Hold, then call back

Learn moreDigital arrest

02 / 07

The fraud moved. The liability is following it.

Book a demoRun a free retrospective

Three figures, each from its publisher: what India’s customers lost to cyber fraud, how much of it was a customer sending her own money, and the date from which RBI’s revised liability framework applies.

India’s rules, dated and sourced

Why fraud still gets through the stack

  • Eight contracts

    Each tool bought, renewed and reviewed on its own.

  • Eight integrations

    Each SDK in your app, each with its own release.

  • Eight places to look

    Your analyst jumps between dashboards to piece one case together.

  • Nobody sees the session

    Each tool sees its own moment. The fraud is in what joins them.

Pick a session. See what DeepID decides, and why.

Switch a signal and watch the policy decide, the customer’s screen change and your FRM record follow.

Policies your fraud team tunes

Try a session
Signals in this session 4 on

Payments policy strictest match wins

  1. An emulator or a tampered appBlock
  2. On a call, with the screen shared or a mule-linked payeeHoldDecides
  3. A payee linked to other accountsCool offMatches
  4. The SIM moved, or a new deviceStep up
  5. Typing unlike the ownerStep up
  6. On a call during a paymentWarnMatches
  7. A first-time payeeWarnMatches
  8. Nothing else matchesAllow
DecisionHold
Your customer seesPayment paused. Your money hasn’t left your account. We’ll call you on +91 •••• 3210.
Your FRM getsHOLD · active call · screen share · first-time payee · mule-linked payeeACTIVE_CALLSCREEN_SHAREDFIRST_TIME_PAYEEPAYEE_MULE_LINKED
Shown in

Decision: Hold. Your customer sees: Payment paused. Your money hasn’t left your account. We’ll call you on +91 •••• 3210.

Build with DeepID. One SDK, every platform.

Sandbox

Wire one moment on test traffic, before any contract.

Get sandbox access
What comes backdecision.json
{  "action": "hold",  "reasons": [    "ACTIVE_CALL",    "SCREEN_SHARED",    "PAYEE_MULE_LINKED"  ],  "message": "Payment paused. Your money hasn’t left your account. We’ll call you on +91 •••• 3210.",  "policy": { "id": "new-payee-payment", "version": 14 }}

SDK and APIs

Android, iOS, React Native, Flutter and web. One call per moment, one decision back.

QuickstartRead the docs
Your app, before the payment is sentpayment.ts
const decision = await deepid.decide("payment.initiate", {  customer: user.id,  payee: payee.id,  amount: { value: 96000, currency: "INR" },  rail: "UPI",}) if (decision.action === "allow") sendPayment()else showDecision(decision.message)

On your own phone

Sign in with one tap. Nothing to type.

Try itLearn more

Settled before the pilot What the SDK reads on the phone, where decisions are processed and how long anything is kept are agreed with your security team in writing before a pilot. Sandbox access comes before any contract. What is agreed first

Built for India’s rules, first. Cited, rule by rule.

India first. The rule map tracks the same six controls across Singapore, Malaysia, the Philippines, Australia, the UAE and Saudi Arabia, each dated and sourced. DeepID supports compliance and never guarantees it.

India and six more markets

  • In force 1 Apr 2026

    One-tap login that tries the network check, the SIM check and Aadhaar consent before any one-time code, and face checks with injection defence where identity is in doubt.

    RBI press release

  • Issued Nov 2025 · deadline reported as 31 Dec 2026

    SIM and device binding at onboarding and re-binding on change, with the number confirmed by the network and device integrity read every session.

    MediaNama, press report

  • Live since Dec 2024 · MoU May 2026

    A payee and mule graph across accounts and devices, with recipient risk scored before the payment is sent.

    Outlook Business, press report

  • Since Jul 2017

    Decision logs with their reasons and the message shown to the customer, and versioned policies, as evidence of what the bank did.

    RBI circular

Proof on your own data, before you commit.

We publish no success rates. You measure DeepID on your own traffic, against criteria agreed in writing before the pilot starts.

The pilot, step by step

Shadow proof to production 14 to 20 weeks

  1. 01Try itYour own phone
  2. 02RetrospectivePayee and mule links
  3. 03Shadow proofYour own customers
  4. 04Limited productionFirst journey live
What you measure

The fraud caught that your controls missed, the genuine customers it would have stopped, the effect on your review queue, and the time each decision takes.

Request a free retrospectiveTry one-tap login

The questions banks ask before they start.

Every question

How do you tell a coerced customer from a busy one?

Not from one signal. A call during a payment is common. A 41-minute call with an unknown number, a shared screen, a payee saved minutes ago and typing in long pauses, all together, is not.

Your fraud team decides which combinations warn, cool off or hold, and sees what each would have done in the shadow proof before anything acts on a customer.

Will it hold genuine payments?

Some, if the policy is strict, which is why the shadow proof comes first. DeepID reads your live traffic without acting, and you compare its decisions with your own outcomes, including how many genuine payments each policy would have held, before you choose how strict to be.

Who talks to the customer?

You do. DeepID returns the action and the words your app shows. Your call-back team, your scripts and your channels stay yours.

Does DeepID replace our FRM?

No. DeepID sends one decision with its reasons to your app and to your FRM, case manager or SIEM, as an input your own rules can use. Teams start with one journey, such as sign-up or login, and keep the rest of the stack they run today.

Is a network check enough to trust a login?

On its own, no, and DeepID doesn’t treat it as the decision. A network check confirms that the SIM for the number is in the phone making the request. A SIM swapped at a store would pass it.

That is why the same decision also reads the SIM and device binding from sign-up, the device itself and, on a new phone, a face check against the one from sign-up. A one-time code proves only that someone received the code, not who typed it.

What does a pilot measure?

What you agree in writing before it starts. Usually four things: the fraud DeepID caught that your controls missed, the genuine customers it would have stopped, the effect on your review queue, and the time each decision takes. The genuine customers stopped is always one of them.

How long before behaviour signals are useful?

It depends on how often your customers use the app, so it is measured on your own traffic during the shadow proof rather than promised up front.

Where is our data hosted, and for how long?

Hosting, data residency and retention are agreed with your security team before the pilot, and sandbox access is available before any contract. The full security pack is shared on request.

Try DeepID before you buy it.

One-tap login on your own number, a sandbox for your engineers, then a pilot on one journey, measured on your own traffic.

Or start with a free retrospective

Book a demoTry one-tap login