Your Customer Does Not Know They Were Denied by a Model
A rejection letter that never mentions a model isn't neutral - it's a disclosure gap. Why silent algorithmic denial is becoming BFSI's next transparency problem.
Read more →Sovereign Stack
AI governance for India's regulators.
Weekly analysis on RBI FREE-AI, DPDP, SEBI circulars, and what they actually mean for compliance officers in BFSI.
RSS feedModel Governance Does Not End at Deployment. That Is Where It Starts.
Most governance effort is spent before a model ships. Here's why the deployment date is the start of the highest-risk phase, not the finish line.
Read more →Your AI Is Only As Compliant As Its Training Data's Paper Trail
Model documentation means nothing if you can't prove what the model was trained on. Why training data provenance is the compliance gap BFSI institutions haven't priced yet.
Read more →You Cannot Govern Models You Have Not Listed
The shadow AI inventory problem: BFSI institutions can't govern models, prompts, and agents nobody registered. Here is why that gap is now a regulatory exposure.
Read more →A Board-Approved AI Policy Is Not a Document. It Is a Set of Decisions.
Most AI policies in Indian financial institutions decide nothing. Here is the test that separates a governance instrument from a well-formatted intention.
Read more →Can You Reconstruct One AI Decision From 90 Days Ago?
RBI's FREE-AI audit framework turns AI governance policy into governance proof. Most institutions can't reconstruct a single AI-assisted decision today.
Read more →Hallucination just became a board-level risk.
RBI's latest draft quietly changes who owns AI hallucinations. What was once considered a product problem is becoming a governance obligation.
Read more →The era of AI policy is ending. The era of AI evidence is beginning.
Inside RBI's draft Guidance on Model Risk Management - what the shift from AI policy to AI evidence means for regulated entities.
Read more →