Niriksha: Health Claims Adjudication using Agentic AI Qorus-NTT DATA Innovation in Insurance Awards 2026

Niriksha is Bajaj General’s Agentic AI solution to modernize health claims adjudication by combining Generative AI with connected data and specialized validation agents. In the current process, adjudicators and technical teams manually review hospital bills, discharge summaries and supporting documents, cross-check policy terms, tariffs and medical protocols, and raise queries with hospitals in discrepancy cases. Niriksha automates these high-effort steps using an agentic architecture: an Intelligent Document Processing (IDP) agent performs OCR and line-item extraction from bills and documents; specialized agents then validate technical rules, bill abuse patterns, medical protocol adherence, tariff deviations and pharmacy anomalies. Discrepant claims are presented in a dedicated Niriksha application with line items highlighted, enabling adjudicators to focus on exceptions and take faster, more consistent decisions.

Concept & objectives: Automate claim intake and document understanding; standardize rule enforcement; detect abuse/overcharging; provide ‘medical copilot’ decision support; and generate analytical data products for downstream monitoring and governance.

Reasons behind: Rising health claim volumes and document complexity require faster, more reliable adjudication with strong compliance. Niriksha is designed to cut turnaround time, improve accuracy and reduce leakage through systematic validations.

State of competition & inspiration: Industry solutions often digitize documents or apply isolated rules. Niriksha was inspired by the ‘agentic’ pattern - multiple collaborating AI agents - so that document understanding, medical reasoning and tariff analytics can be composed into an end-to-end adjudication workflow.

Departments involved: IT Innovation, data engineering and application engineering team, Health Claims comprising of  Medical team, Technical assessors.

Main results so far: Niriksha is positioned to deliver ~2 - 3% savings on cashless claim amounts, reduce adjudication TAT, improve IRDAI compliance and enable fraud/abuse detection.