Question 1

What is Didit?

Accepted Answer

Didit is **infrastructure for identity and fraud** — the platform we wished existed when we were building products ourselves: **open, flexible, and developer-friendly**, so it works as a real part of your stack instead of a black box you integrate around.

One API covers verifying people (**KYC**, *know your customer*), verifying businesses (**KYB**, *know your business*), screening crypto wallets (**KYT**, *know your transaction*), and monitoring transactions in real time — on a stack built to be:

- **Fast** — sub-2-second p99 on every session
- **Reliable** — in production with **1,500+ companies** across **220+ countries**
- **Secure** — SOC 2 Type 1, ISO 27001, GDPR-native, and formally attested by Spain's financial regulator as safer than verifying someone in person

The footprint underneath: **14,000+ document types** in **48+ languages**, **1,000+ data sources**, and **200+ fraud signals** on every session.

Question 2

How fast is the door-side auth?

Accepted Answer

Sub-2-second end-to-end. Camera captures, passive liveness, 1:N match against the bound roster — door opens before the user has stopped walking. Faster than a fob tap.

Question 3

Anti-spoofing — what about masks and printed photos?

Accepted Answer

Passive liveness with **iBeta Level 1 PAD**. Defends against printed photos, video replay, 3D masks, deepfake projection, and morph attacks. Higher security than a physical badge — badges can be stolen, faces with liveness cannot.

Question 4

What use cases is this built for?

Accepted Answer

Offices and co-working buildings (replace the keycard), gyms and member clubs (replace the wristband), event ticketing (replace the QR), residential buildings (replace the intercom), hotels (replace the room key), transit and stadiums. Anywhere a physical credential gates entry.

Question 5

Hardware — what cameras are supported?

Accepted Answer

Any IP camera ≥720p with reasonable light. Our SDK runs inference cloud-side, so no specialised on-prem appliance is required. Existing CCTV + access-control vendors integrate via REST webhook to call `/v3/face-search/`.

Question 6

Per-entry price?

Accepted Answer

**$0.33 one-time KYC enrolment per user** (ID + Passive Liveness + Face Match + IP Analysis), then **free per door auth** — Face Search 1:N is bundled free on every Didit account. 500 verifications free every month. Predictable monthly invoice — no surprise surcharges, no minimums.

Question 7

Privacy — can I keep face templates on-prem?

Accepted Answer

Production templates live in the EU by default on AWS, encrypted under a separate Customer Master Key. Enterprise contracts can request alternative regions for jurisdictions whose regulators require it. Default retention is indefinite but you control it — between **30 days and 10 years per application** — and you can delete any individual session at any time from the dashboard or the API.

Question 8

What if a user fails or is not enrolled at the gate?

Accepted Answer

The signed match result lets your gate logic branch. Match → unlock. No match → route to staff or to a tablet for fresh enrolment (one $0.33 session). Already-banned → block, log, alert. Same Workflow Builder pattern as a KYC flow.

Question 9

Multi-site — can I run one roster across many doors?

Accepted Answer

Yes. The face index is per application, not per terminal. Enrol once at any door and the user is recognised at every other door on the same application — useful for multi-site offices, multi-gym chains, transit corridors, hotel chains, and festival circuits.

Question 10

Does it support disability accommodation?

Accepted Answer

Yes. The hosted capture screen runs at WCAG 2.2 AA. For users who can't self-capture, staff can run an assisted-enrolment session from a tablet. Card-and-PIN or fob fallback stays available — face-on-door is an additive credential, not a replacement.

Question 11

Is the door auth GDPR-compliant?

Accepted Answer

Yes when consent and lawful basis (Article 9 GDPR) are explicit and the template is stored under your control. Didit ships the consent UX and the audit trail. The stack is iBeta Level 1 PAD certified, SOC 2 Type 1, ISO/IEC 27001:2022, and formally attested by Spain's Tesoro / SEPBLAC / CNMV as safer than verifying someone in person.

Question 12

How fast can I integrate face-on-door?

Accepted Answer

**60 seconds** to a sandbox account at [business.didit.me](https://business.didit.me) — no credit card.

**5 minutes** to a working enrolment + door-side match via Claude Code, Cursor, or any coding agent through our MCP server.

**A weekend** to a production-ready integration that wires existing CCTV or access-control vendors. Step-by-step guide at [docs.didit.me/integration/integration-prompt](https://docs.didit.me/integration/integration-prompt).

translation_v1.solutions.accessControl.hero.theFaceIsBlue

translation_v1.solutions.accessControl.noFobNoQr

From fob-tap to face-at-the-door in four steps.

Enrol once at any door

Wire the door to Face Search 1:N

User walks up to the door

Audit + monitor

translation_v1.solutions.accessControl.sectionintro.033EnrolmentBlue

One $0.33 session binds a face to an authorised user.

The user walks up. The door finds them.

Defeats printed photos, replays, masks, deepfakes.

One face. Every door on the application.

Entry. Re-entry. Block. Configurable.

Face-on-door adds. Nothing breaks.

Open a new country in one click. We do the hard work.

Common questions

Face Search 1:N

Passive Liveness

Biometric Authentication

Face Match 1:1

Biometric event ticketing

Biometric payments

Infrastructure for identity and fraud.