Face Verification

The right face, on the right CPF.

Liveness detection and selfie-to-CPF-holder matching, delivered through a hosted link. No SDK, no changes to your application — on the same query API you already use.

The flow is designed together with our team.

your brand verificacao.sua-marca.example
  • Liveness
  • Face match against the CPF
Verification approved
No SDK
Delivered through a hosted link
White-label
A page carrying your brand
1 webhook
The verdict straight into your backend
130+
Queries on the same key
Unico
Biometric validation
How It Works

Four steps between sign-up and verdict

None of it requires touching your application. Verification happens outside of it and comes back done.

  1. 1

    Your application calls the API

    One request with the holder's CPF and your webhook URL. The response carries the verification link.

  2. 2

    The link reaches the user

    Send it through the channel you already use — email, SMS, WhatsApp or inside your own flow. If you prefer, we send it for you.

  3. 3

    They take a selfie on the hosted page

    Passive liveness: the user simply looks at the camera. The page is white-label — your brand, your colours, your domain.

  4. 4

    You get the verdict by webhook

    Approved or rejected, with the liveness result and the match against the CPF holder. Your application decides what happens next.

face-verification
// 1. your application creates the verification
$ curl -X POST -H "X-API-Key: SUA_CHAVE" \
  -d '{"cpf":"12345678900","callback_url":"https://sua-app.com/hook"}' \
  "app.fontedata.com/api/v1/verificacao-facial"

// 201 Created — returns the hosted link
{
  "id": "vf_9f2c41",
  "status": "pending",
  "url": "https://verificacao.sua-marca.example/vf_9f2c41"
}

// 2. the user takes the selfie on your branded page
// 3. POST to your webhook — the verdict
{
  "id": "vf_9f2c41",
  "status": "approved",
  "cpf": "123.***.***-00",
  "liveness": true,
  "face_match": true
}
status: approved APPROVED
Custom Flows

1:N at the door, 1:1 on every sensitive transaction

Every verification starts from a design made with you: how many attempts, what fails, what goes to human review.

1:1 · on request Transactional: is this the same user as always?

The selfie of the moment is compared against the photos you already have of that user — the reference captured at onboarding. It re-authenticates someone who is already your customer, without sending them through sign-up again. Custom flow: we design it with you.

  • Risky login and device change
  • Transaction or withdrawal approval
  • Changes to registration data
And whatever else your flow calls for

Decision rules, number of attempts, routing to human review, branching by outcome, automatic link resend. Tailored flows are the rule, not the exception.

Why Now

Identity fraud has reached industrial scale — and the law has caught up

In twelve months, 24 million Brazilians were victims of Pix-related scams, with R$ 29 billion in losses, according to a Febraban survey with Datafolha. This is not artisanal fraud: Serasa Experian recorded 10.89 million fraud attempts between January and September 2025 — one every 2.2 seconds.

The technology on the other side has accelerated too. Sumsub reports a 126% rise in deepfakes in Brazil from 2024 to 2025. Once a face can be synthesised, confirming that a real person is in front of the camera stops being a refinement and becomes a requirement.

The regulatory bar rose alongside it. The ECA Digital act (Law 15,211/2025), in force since 17 March 2026, requires effective age verification — self-declaration no longer counts — under fines of up to 10% of revenue, capped at R$ 50 million. In the betting market, Ordinance SPA/MF 722/2024 already requires facial recognition when a bettor registers.

R$ 29 bn
lost to Pix-related scams over 12 months
Febraban/Datafolha
10.89 M
fraud attempts from Jan to Sep 2025 — one every 2.2 seconds
Serasa Experian
+126%
in deepfakes in Brazil from 2024 to 2025
Sumsub
R$ 50 M
cap on the ECA Digital fine, or 10% of revenue
Law 15,211/2025
Who It Is For

Where face verification pays for itself

Betting and iGaming

Ordinance SPA/MF 722/2024 requires facial recognition when a bettor registers. The hosted link slots into onboarding without rewriting the app.

Fintechs and PSPs

An account opened with someone else's identity is the front door to a mule account — and the start of a chargeback dispute. Liveness at sign-up sharply reduces that risk.

Platforms under ECA Digital

Social networks, online course platforms, streaming and app stores need effective age verification. Self-declaration no longer counts.

Marketplaces

Sellers, couriers and high-value buyers: confirm who is on the other side before releasing the payout.

Rental and mobility

Car, property, equipment. Confirming the holder's face before handing over the asset costs less than recovering it afterwards.

HR and remote hiring

Onboarding without a face-to-face meeting: confirm that whoever signs is whoever was interviewed.

The Difference

Approved? Now query.

An approved verification is not the end of the flow — it is the beginning of it.

Face verification confirms that a real person exists and that they hold the CPF you sent. What it does not tell you is who that person is. At FonteData that is one call away: once the verification is approved, your application fires whatever queries you want on the same CPF — registry, criminal record, PEP, credit score, vehicle — on the same key, the same billing and the same dashboard.

Verification and a data dossier in the same API — and in the same callback.

Prices come from the live catalogue. You choose which queries join the flow — none of them is mandatory. See the full price list →

Transparency

What the verification delivers — no fine print

Ownership

We validate that the selfie belongs to the holder of the CPF you provided.

Liveness

Liveness detection to confirm there is a real person in front of the camera, not a photo or a video.

Age verification

It helps meet the ECA Digital age-verification requirement, which bars self-declaration.

Third-party identity

It reduces the risk of an account being opened with someone else's identity.

Sensitive data, handled as such

Biometric data is sensitive personal data under Brazil's LGPD. It is collected for a specific purpose, with a defined retention period and no reuse for anything else.

Questions

Frequently asked questions

See the full documentation →

Let's design your verification flow

Talk to the commercial team and build the flow with us — from 1:N at sign-up to the full dossier in the same callback.

Already a FonteData customer? Verification runs on the same key.