Video KYC for Insurance: IRDAI Guidelines vs Ground Reality in 2026

Posted by

Ask most insurtech product teams what Video KYC for insurance looks like and you’ll get a confident answer. Ask them how many agents actually complete a VBIP session without a drop-off, connectivity issue, or a compliance gap, and the answer gets quieter.

That gap — between what the guidelines say and what actually happens — is where most insurers quietly struggle. And in 2026, with IRDAI’s focus on digital onboarding only intensifying, it’s a gap worth understanding properly.

What IRDAI Actually Mandated

IRDAI’s Video Based Identification Process, or VBIP, was introduced as an alternative, paperless KYC channel for life, general, and standalone health insurers. It is defined as a real-time, consent-based audio-visual interaction between an authorised insurer representative and the customer, carried out to obtain identification information and KYC documents needed for client due diligence.

The process isn’t just a video call. Insurers must carry out software and security audits of the application before rolling it out, ensure end-to-end encryption using minimum standards such as AES 256, host the application within India, and have the app and related APIs undergo security testing through a CERT-In empanelled vendor.

The authorised person conducting VBIP must be specifically trained for it, and insurers are required to maintain details of every such person facilitating the process. Insurers must also ensure Aadhaar numbers are redacted or blacked out as per extant PML Rules.

On paper, this is a thoughtful framework. It treats video KYC as equivalent to face-to-face interaction, which is significant — it removes the need for physical visits while maintaining compliance weight. From January 2023, IRDAI mandated KYC procedures across all classes of life, general, and health insurance, with Aadhaar-based KYC, digital KYC, and Video KYC all recognised as accepted forms.

So far, so good. The problem is what happens when this meets reality.

Ground Reality: What’s Actually Breaking Down

The connectivity problem nobody talks about

VBIP requires a seamless, real-time audio-visual interaction. That’s a reasonable requirement in Bengaluru or Pune. In a Tier 3 town in Bihar or a semi-urban pocket in Odisha, it’s a gamble. Call quality degrades, video freezes mid-session, and the authorised agent has to restart the process from scratch.

There’s no provision in the IRDAI guidelines for asynchronous video KYC — the kind where a customer records a consent video that’s reviewed offline. Every session has to be live. For insurers trying to onboard customers in low-connectivity geographies, this creates a hard ceiling on reach.

Trained authorised persons: more bottleneck than safeguard

The guideline that VBIP must be conducted by an authorised, specifically trained representative is well-intentioned. In practice, it means that scaling video KYC is directly constrained by how many trained agents an insurer has available at any given time.

Peak hours — evenings, weekends, the last few days of a quarter when agents are pushing policy closures — create a queue. Customers drop off. Agents rush sessions to clear backlogs. Neither outcome is what the regulator had in mind.

The geo-tagging expectation vs the infrastructure

IRDAI guidelines ask for geo-tagging and a live photograph of the customer during the VBIP session. In theory, this strengthens fraud prevention. In practice, insurers using third-party apps that haven’t been built specifically for VBIP often find that geo-tagging data is inconsistently captured, particularly on older devices where location permissions behave unpredictably.

This is an audit problem waiting to happen.

Consent management in regional languages

IRDAI requires informed consent before a VBIP session begins. Most insurers collect it via a checkbox or a brief verbal acknowledgment on the call. But in a country where a meaningful portion of insurance buyers are first-generation digital users, a consent mechanism designed for English-literate urban customers isn’t serving its purpose. Consent in spirit requires comprehension, not just a click.

Where the Regulation Is Ahead of the Technology

The IRDAI guidelines were written to accommodate emerging technology, and they do so reasonably. They encourage the use of AI to ease the video KYC process. But the reality is that most insurers aren’t running AI-assisted liveness detection or deepfake screening on their VBIP flows. They’re running video calls with a trained agent watching closely.

That’s not inherently wrong — a trained human eye catches things algorithms miss. But as synthetic identity fraud becomes more sophisticated in 2026, a purely agent-dependent VBIP process starts to look underpowered. The guideline permits AI; the infrastructure to run it reliably isn’t uniformly in place yet.

What Good Video KYC for Insurance Looks Like in Practice

The insurers doing this well share a few common traits. They’ve built or integrated a VBIP solution that was designed specifically for the insurance use case — not adapted from a generic video conferencing tool. Their agent training isn’t a one-time onboarding module; it’s updated regularly as fraud patterns evolve. They’ve accounted for drop-off at every step and built re-engagement logic into their flow.

They also treat geo-tagging, session recordings, and audit trails not as compliance checkboxes but as operational data — something that gets reviewed when things go wrong, and that informs how the process is tightened over time.

And critically, they’ve tested their VBIP flow specifically in low-connectivity environments before rolling it out to semi-urban and rural customers. That’s not a technical afterthought. For an insurer with serious pan-India ambitions, it’s a precondition.

The Bottom Line

IRDAI’s VBIP framework is directionally sound. It recognised early that video-based identification, done right, is as robust as physical verification — and far more scalable. The guidelines on encryption, audit trails, trained agents, and geo-tagging reflect genuine thinking about what could go wrong.

The gap isn’t in the regulation. It’s in execution. And for insurers serious about using video KYC for insurance as a real growth lever — not just a compliance checkbox — closing that gap requires investment in infrastructure, agent training, and technology that matches the standard the regulator has set.

The framework is there. The question is whether your implementation is.

Leave a Reply

Your email address will not be published. Required fields are marked *