For modern financial institutions, the "Welcome" mat has become a critical efficiency checkpoint in the customer lifecycle. 

However, a significant friction point remains. As banks race to digitize, many have implemented security measures that inadvertently erode user acquisition. The result is a paradox: the harder an institution tries to secure the front door using outdated tools, the fewer legitimate customers actually enter. 

According to Fenergo’s Financial Crime Industry Trends 2025, the situation is pressing: 70% of financial institutions globally lost clients in the last year specifically due to slow or inefficient onboarding processes . This represents not just a user experience challenge, but a measurable operational loss. 

The critical question for financial leaders is not whether to use biometrics, but how to orchestrate them effectively. The industry often debates "Active vs. Passive," but this is a false dichotomy. The real issue is the indiscriminate application of first-generation, gesture-heavy active liveness that frustrates users without necessarily mitigating modern fraud risks. 

Here is how a risk-based approach and superior technology change the equation. 

The Friction Point: When Legacy Tech Meets Modern Users 

To optimize conversion, we must diagnose the bottleneck. The modern consumer operates in an instant-attention economy. When a potential client downloads a banking app, they expect a seamless flow. 

Legacy Active Liveness solutions, those designed in the pre-GenAI era, often treat every user as a high-risk subject. They demand complex "digital gymnastics": Blink twice. Turn left. Move the phone closer. 

While these instructions may serve a purpose in specific contexts, applying them as a blanket requirement creates unnecessary barriers. 

• The Cost of Time: Research from Deloitte indicates that 38% of customers abandon account creation simply because the process takes too long.

  
  

• Operational Impact: When a legacy system takes 15 seconds to process a video stream only to reject a legitimate user due to poor lighting or a trembling hand, the cost of acquisition (CAC) spikes. Industry data suggests nearly 68% of consumers have abandoned a financial application due to such technical friction.

The conclusion is clear: Indiscriminate friction erodes conversion. 

The Security Challenge: Gestures vs. Generative AI 

Beyond UX, the reliance on first-generation active liveness faces a security challenge. In an era where synthetic fraud is projected to cost $40 billionannually by 2027 , relying solely on simple gestures is no longer a guarantee of security. 

The threat landscape has evolved to "Puppetry" and "Injection": 

• The Puppet Master: Generative AI tools allow fraudsters to perform "Puppetry" attacks, where a deepfake avatar mimics the fraudster's expressions in real-time. A legacy system asking a user to "smile" may be fooled by a deepfake that smiles instantly.

  
  

• The Injection Vector: Sophisticated actors bypass the physical lens entirely using virtual camera software to inject synthetic video directly into the data stream. Legacy systems looking for optical artifacts (like screen glare) may fail to detect this, as there is no physical presentation.

This creates a scenario where banks may be subjecting legitimate users to high friction while leaving the door ajar for sophisticated AI-driven attacks. 

The Strategic Solution: Risk-Based Orchestration 

At Oz Forensics, we believe the future of onboarding lies in Risk-Based Orchestration. It is not about choosing one liveness method over another, but about modernizing biometric assurance so that security is driven by attack detection, not user friction. 

Our platform empowers institutions to deploy a dynamic defense strategy: 

1. Secure, Low-Friction Capture 

Capture that is done quickly and safely through highly secure multi-frame liveness, removing unnecessary user effort while maintaining strong protection against modern attacks. 

2. Liveness That Detects Modern Attacks 

Liveness effectiveness is no longer defined by gestures, but by the ability to detect deepfakes, puppetry, and synthetic manipulation, without relying on user friction. 

3. The Invisible Shield: Injection Attack Detection (IAD) 

Regardless of whether you use Active or Passive, the integrity of the data stream is non-negotiable. Our Injection Attack Detection (IAD) works in tandem with our biometric checks. It analyzes the video stream's metadata and signal integrity to ensure the footage originates from a physical camera sensor. This aligns with the European technical specification CEN/TS 18099, as verified by independent BixeLab testing, providing defense-in-depth against synthetic injection . 

Conclusion: The Right Tool for the Right Risk 

The goal of modern governance is to balance security posture with business growth. 

Your customers are ready to join. The technology stack should facilitate their entry, not hinder it. By moving away from legacy, one-size-fits-all biometric checks and adopting Oz Forensics’ orchestrated approach, financial institutions can stop deepfakes and block injection attacks while ensuring a frictionless path for legitimate revenue. 

Optimize Your Onboarding Without Compromising Security  
Learn how modern liveness resolves the friction paradox.