“AI will not replace people, but people using AI will replace those who don’t.” - Paul Daugherty, Chief Technology & Innovation Officer, Accenture
AI’s growing footprint in public procurement
India’s procurement system, contributing nearly 20–22% of GDP, is the backbone of government spending. From highways to healthcare, efficiency and transparency are critical. Yet, officers still navigate thousands of documents, cross-checking bids and verifying certificates under tight deadlines. While platforms like GeM have digitised the process front-end, the real challenge lies in evaluating and interpreting the extensive backend data.
Automating compliance through intelligent systems
AI-powered tools such as GenAI are now transforming this space by automating tender scrutiny. Using Optical Character Recognition (OCR) with semantic parsing, AI can scan and extract data from massive documents in minutes. Systems ingesting bids from portals like CPP or IREPS can flag anomalies, verify eligibility, and highlight non-compliance, helping officers focus on decision-making instead of routine checks.
Structuring evaluation for transparency
Machine-Readable Bid Annexures (MRBA) and Clause-to-Evidence Mapping (CEM) enable structured uploads of tender clauses and related financials. AI algorithms highlight compliance gaps and generate comparative statements, ensuring every tender is evaluated transparently. Risk and Exceptions Registers (RER) further visualise discrepancies, embedding accountability while reducing manual errors.
Policy integration and benchmarking
By integrating 1,614 data layers across 44 ministries and 36 states, India’s procurement ecosystem under PM Gati Shakti creates a unified evaluation system. This allows AI to compare bids across similar projects, regions, and timeframes, ensuring fairness and accuracy in benchmarking.
The road ahead for AI-driven procurement
AI in procurement won’t replace human judgment but will streamline processes. Officers can use AI-assisted pre-tender analysis to standardise evaluations and support faster decisions, gradually building institutional memory for transparent governance.

COMMENTS