We first objectively questioned the legacy narrative that credentials and cryptography are deterministic and can ‘identify’ users. They cannot. There is always the chance the user holding the device or credentials isn’t the person they were issued to, it’s a cloned device, or an impostor had new keys issued to a parallel device.
Adding factors reduces impostors using credentials, but they never positively identify the correct user. This led to understanding biometrics’ critical role and the need for 3D Liveness Detection. We aren’t authenticating devices: proving Liveness means the physical user is present at the exact moment biometric data is captured with extremely high confidence, remotely, and unsupervised.
Further, we deploy software, not a cloud service, that runs inside our customers’ firewalls, so no biometric data is sent to FaceTec. This privacy-by-design ensures adherence to ever-changing, global privacy frameworks.
We were the first to test with a NIST-certified, independent lab, but testing was – and still is – indefensibly inadequate. It’s imperative to know where and how our systems might be attacked in the real world before any customers do. The only global, public, ongoing spoof bounty program is the ultimate way to learn in real-time what the state of the art in attacks are. Launched in October 2019, and after more than 150,000 attacks, we’re still the only vendor with a spoof bounty program. Most don’t even have publicly available demos!
We invite attacks, believing if we aren’t strong enough to defend our own bounty then we shouldn’t sell to other companies. In fact, no Liveness vendors should without the confidence their Liveness can stand-up a spoof bounty program. NIST 2D Liveness PAD test results reveal why those vendors can’t have a program. All 42 2D vendors were spoofed in every artefact category by basic PAD attacks.
FaceTec hasn’t paid a bounty in over three-and-a-half years and has had zero successful spoofs accepted in recent 3D Liveness NIST-Lab PAD testing.
First, three technology characteristics make our systems globally deployable. Our software is exceptionally efficient, universally compatible with billions of smart devices and webcam-enabled computers, and runs flawlessly on all major operating systems. Our AI models only require a 0.3 MP-plus camera, an approximately 4 MB device SDK, and a 350 KB 3D FaceScan payload for the highest-confidence liveness detection and biometric matching. This enables deployments in markets with very inexpensive devices and slow networks, allowing us to process over 2.4 billion-and-growing liveness checks annualized, even in difficult rural areas and on cruise ships.
Secondly, our face data training set is arguably the most diverse. Volunteers providing face data from multiple devices in multiple, diverse scenarios allow human face constants and variables to be learned from 180-plus countries, ensuring our technology isn’t hindered by AI models trained on homogeneous 2D face photos.
Importantly, our Liveness and biometric face matching are 3D. 2D cameras struggle to accurately represent the human face’s 3D shape because of inconsistent perspective distortion from flattening 3D features. 2D ‘guesses’ about how faces are really shaped can increase errors. 2D must rely on less consistent characteristics, including skin tone and color saturation. Darker skin reflects less light than lighter skin, effectively reducing measurable data, increasing match error rates. 2D can affect demographic results from young people, females, and wearing makeup or glasses.
FaceTec provides self-hosted software with no third-party data privacy and security concerns. FaceTec doesn’t require end-user biometric data to be sent to us, fundamentally different from centralized, cloud-based processors, like most KYC services. In addition to privacy-by-design architecture, we only charge for 3D Liveness Checks. All other IDV/KYC features are free for unlimited use, including photo ID scans, text OCR, NFC chip reading, barcode scans, and 1:N for de-duplication and bad-actor face bans. Our software is a fraction of the fraud and brand damage costs customers could experience without it.
Resistance to 3D Liveness-proven biometrics, or those handcuffed to legacy thinking, should study the recent attacks on Okta and MGM. They could only have been stopped by biometrically exposing the impostors who had the right credentials. If those don’t change minds, the coming tsunami of infinitely scalable GenAI-powered deepfakes trained as social engineers will. The legacy cybersecurity establishment can no longer afford to ignore that they do not provide positive user identification. Each hoop added only slightly reduces the impostor risk, creating a false sense of security. Liveness-proven biometrics are necessary, and FaceTec will continue to provide the world’s best privacy-preserving face biometrics.
This editorial was initially published in the Emerging Technologies and Trends in Identity Verification, KYC, and KYB Report 2024. The report dives into the latest practices and technologies that enable financial institutions and regulated entities to reduce fraud, build trust, navigate evolving regulatory and compliance requirements, and cut operational costs. You can download your free copy here.
Founder and investor, Kevin builds businesses that create global value. Most noteworthy, FaceTec is a leading biometric cybersecurity firm he founded in 2013 that enables frictionless biometric account security on billions of existing devices. A foremost expert in anti-spoofing technology, he holds numerous presentation attack detection-related patents.
US-based with staff in Brazil, Portugal, Mexico, and Singapore, FaceTec is the world's leading provider of 3D Liveness and Face Verification software, providing well over two billion 3D Liveness Checks annually, bringing remote identity proofing to a much higher level of security.
Every day we send out a free e-mail with the most important headlines of the last 24 hours.
Subscribe now