Model Risk Management for Indian Banks and NBFCs

TL;DR A credit scoring model approves loans every hour of every working day. It was validated once, at launch, eighteen months ago. Since then the borrower mix has shifted, a new product line has fed it data it never saw in training, and its accuracy has slipped by a few points a quarter. Nobody flagged […]
Why Human-in-the-Loop Isn’t Optional Anymore: Building Kill Switches Into AI Banking Compliance

TL;DR A collections model starts flagging borrowers it shouldn’t. A fraud model’s false-positive rate creeps up over a weekend and nobody notices until Monday. The model risk officer’s first question in a room like this is never “do we have a model inventory.” Every regulated entity has one of those by now. The real question […]
Stress Testing AI Models: What RBI Expects Before You Deploy

TL;DR Stress Testing Your Balance Sheet and Stress Testing an AI Model Are Two Different Exercises When a risk officer at an Indian bank hears “stress testing,” the reflex is capital adequacy: RBI’s macroprudential scenarios, liquidity coverage under shock conditions, provisioning against a stressed balance sheet. That discipline is well understood, well documented, and audited […]
Model Risk Management for NBFCs: A Configuration Guide for Upper-Layer Compliance

TL;DR NBFC-UL Compliance Teams Assume the MRM Draft Gets Stricter at Their Layer Most NBFC-UL risk and compliance teams reading RBI’s June 2026 model risk management draft guidance make a reasonable assumption: because their entity carries the highest regulatory burden under Scale-Based Regulation, the new MRM requirements must scale up with it. Somewhere in the […]
What Your Model Risk Committee Should Actually Be Reporting to the Board

TL;DR Most Model Risk Committees Are Reporting Against a Standard That Doesn’t Exist Yet Ask a Model Risk Management Committee chair at an Indian bank or NBFC what their board report should contain, and you will get a confident answer built on habit rather than rule. That reflects where the current regulatory landscape actually stands. […]
What Counts as an AI Model Under RBI’s New Guidance, and How to Keep an Audit Trail

TL;DR: RBI’s Draft Guidance Defines “Model” by Substance, Not by What Your Team Calls It On June 24, 2026, the Reserve Bank of India released a draft guidance document (prid=63006) for public comment, with the comment period closing July 24, 2026. It is a draft. It is not a Master Direction, and it has not […]
How Often Should You Revalidate a Model? RBI’s Answer for NBFCs

TL;DR RBI Has Not Set One Number for Model Revalidation Frequency If you searched for this because a risk committee member asked “how often does RBI actually require us to revalidate,” the honest answer is that RBI has not published a single number. The Reserve Bank of India released draft guidance on model risk management […]
Model Governance Software for Indian Banks: What the Category Actually Offers

TL;DR Vendor Certification Does Not Reduce Your Validation Burden Under RBI’s Draft Guidance A model comes with a vendor certificate. The vendor ran its own testing, its own bias checks, its own performance benchmarks, and signed off. For most global model governance platforms, that certificate is treated as a meaningful input into the bank’s own […]
Model Risk Tiering: Why Not Every AI Model Needs the Same Scrutiny

TL;DR A model risk officer at an NBFC gets asked by an examiner which of the institution’s forty deployed models get the deepest scrutiny. There is no RBI-published answer yet, only three named inputs and a deadline that keeps moving. The Three Factors RBI’s Draft Actually Names RBI’s June 24, 2026 draft guidance (“Guidance on […]
What RBI’s Model Risk Management Guidance Actually Means for NBFC AI Teams

TL;DR An NBFC’s model inventory usually lives in three places at once: a spreadsheet the risk team maintains, a vendor scorecard from whichever bureau or fintech partner supplied a scoring model, and someone’s memory of what is actually running in production this quarter. Ask that team to produce a validation trail for one specific model, […]

