If your KYC relies on "standard" liveness detection, you aren't just behind the curve. You’re in the dead zone
- Feb 24
- 2 min read
For years, digital onboarding focused on convenience, with measures such as “blink,” “turn your head,” or “smile at the camera” designed to prevent simple spoofing attacks, like printed photographs or static images. These checks, often referred to as standard liveness detection, are increasingly inadequate against today’s threats.
Modern fraudsters are exploiting three critical gaps:
Injection Attacks: Criminals use virtual cameras to inject deepfake video streams directly into the system, bypassing the physical camera entirely. To a standard check, these appear perfectly legitimate.
Synthetic Identities: By blending stolen PII with AI-generated visuals, fraudsters create "Frankenstein" profiles that look real to both credit bureaus and legacy identity checks.
Trust Erosion: As friction increases for real customers, AI-powered fraudsters find the gaps. This undermines the foundation of institutional trust.
The consequences are significant. Beyond financial loss, organisations face reputational damage, compliance risk, and operational strain.
Deepfake Implications for South African Institutions
With deepfake fraud rising by over 1200% (TransUnion 2025 H1 Africa Digital Fraud Trends Report), financial institutions can no longer treat identity verification as a procedural checkbox. A failure to upgrade verification technology risks:
Increased financial loss from undetected fraud.
Customer attrition due to excessive friction.
Exposure to regulatory and compliance scrutiny.
Digital identity assurance must now be treated as strategic infrastructure, combining technological sophistication, regulatory alignment, and maintaining a user-centric design.
Closing the Gaps in Digital Onboarding Security
For most institutions, digital onboarding has transitioned from a competitive edge to a glaring vulnerability. You’ve moved your KYC online to save time, but in doing so, you’ve entered the Digital Onboarding Dead Zone and an awkward boardroom meeting to discuss millions in financial losses.
The solution is not merely adding another step to the onboarding process. Effective digital identity verification now requires forensic-level analysis and zero-trust principles. This includes:
Analysis beyond surface-level liveness to detect AI-generated streams or manipulated data.
Cross-validation of identity claims with device provenance, behavioural signals, and contextual intelligence.
Systems that are designed to achieve high detection rates while remaining equitable across demographic groups.
SprintHive’s independent testing demonstrates the effectiveness of such approaches. In March 2025, an evaluation commissioned by the U.S. Department of Homeland Security confirmed that our leading forensic biometric system achieved 100% fraud detection with zero bias across age, race, gender, and skin tone categories.
These results highlight that SprintHive’s identity verification technology maintains both security and fairness, an essential consideration in regulated markets.
SprintHive: The Architecture of Online Absolute Certainty
The time to evolve beyond “standard” liveness checks is now. Institutions that do will not only strengthen their defences, but they will also future-proof their digital trust foundations.
The fraud is fake. The losses are real. Let’s fix that. Book a demo with one of our experts sales@sprinthive.com, or visit www.sprinthive.com
