A liveness check (also called liveness detection or liveness test) is an automated biometric verification method that determines whether the person in front of a camera is a real, live human being rather than a photograph, a recorded video, a mask, or an AI‑generated deepfake. Financial institutions, fintech platforms, and crypto card issuers commonly use liveness checks during Know Your Customer (KYC) onboarding to confirm that the person submitting an identity document is physically present and that their face matches the ID photo in real time.
Key points / Quick facts
- A liveness check confirms that a real person is present during verification – not a photo, video, mask, or deepfake.
- It is widely used by regulated financial services, including crypto card issuers, to support KYC and anti‑money laundering (AML) compliance – though not every programme employs the same methods.
- There are three main types: active (user performs an action), passive (no action required), and hybrid (combines both).
- Liveness checks are separate from facial recognition – facial recognition asks who someone is; liveness asks whether the input is real.
- The time and friction vary by method: passive checks are typically faster and require no user action, while active checks involve prompts like blinking or turning the head.
What is a liveness check?
A liveness check (or liveness detection) is an automated biometric verification method that confirms whether the person in front of a camera is a real, live human being – rather than a photograph, a recorded video, a mask, or an AI‑generated deepfake. Financial institutions, fintech platforms, and crypto card issuers use it during KYC onboarding to verify that the person submitting an identity document is physically present and that their face matches the ID photo.
Liveness detection emerged because facial recognition alone is no longer sufficient to prevent fraud. A standard selfie check assumes that what the camera sees is genuine – but fraudsters can submit printed photos, replay recorded videos, or use AI‑generated faces to create synthetic identities. Liveness detection closes this loophole by asking not just “Does this face match the ID?” but also “Was this face captured live right now?”
How a liveness check works
A liveness check processes camera input through several stages, from capture to classification.
- Stage 1: Input capture. A camera captures a still image or short video. In active mode, the user follows prompts – blinking, turning the head, or smiling – making the input harder to replicate with a static spoof.
- Stage 2: Biometric signal analysis. The system extracts signals that indicate a live person produced the input – skin texture, micro‑reflections, natural eye movement, and depth cues that flat images cannot replicate. This analysis runs in real time.
- Stage 3: Machine learning classification. The extracted signals are fed into a model trained on real faces and known spoof types – printed photos, video replays, 3D masks, and AI‑generated faces. The model compares the input against those patterns and assigns a confidence score.
- Stage 4: Liveness decision output. The system returns a verdict – live or spoof – along with a confidence score. A failed verdict stops the verification process before document comparison or identity scoring begins.
Why a liveness check matters for crypto cards and payments
In the crypto and fintech space, liveness checks are especially important for several reasons.
First, regulated crypto card issuers are subject to strict KYC and AML requirements. Liveness checks are now a common industry practice – typically requiring users to provide a government‑issued ID and take a live selfie that the system matches to the ID photo. Without this step, issuers would be more vulnerable to synthetic identity fraud.
Second, because crypto assets are often irreversibly transferred and may be volatile, the stakes of identity fraud are particularly high. A liveness check helps ensure that the person opening an account and accessing a credit line is who they claim to be – reducing the risk of unauthorised account creation, money laundering, and fraudulent spending.
Third, liveness checks protect users themselves. If someone steals a copy of your ID, they cannot simply use a photo of your face to open an account – the liveness check will detect that the input is not a live person and reject the attempt.
For crypto card users, the liveness check is typically one of the steps in the onboarding process. Modern passive technology can complete verification quickly and with minimal friction, making enhanced security more convenient than traditional manual checks.
Types of liveness checks
- Active liveness detection: Requires users to perform a specific action – blinking, smiling, turning their head, or reading a random number. Highly reliable against basic spoofs but creates more friction and takes longer than passive methods.
- Passive liveness detection: Runs automatically during a normal selfie capture – no prompts or actions required. Analyses skin texture, light reflection, depth, and micro‑movements invisible to the naked eye. Generally faster and more seamless for the user.
- Hybrid liveness detection: Combines passive analysis with selective active challenges. A passive check runs first – fast and seamless. If the result is uncertain or the user is flagged as higher risk, an active challenge is triggered. Offers comprehensive protection while maintaining a smooth experience for most users.
Understanding the type of liveness check used helps users know what to expect during onboarding. A passive check feels invisible – just a quick selfie. An active check requires following a few simple prompts. Either way, the goal is the same: to confirm you are a real person and protect your identity from fraud.
