Discuss how Artificial Intelligence is reshaping credit underwriting for MSMEs in India. What are the associated risks and regulatory challenges?

MSMEs contribute over 30% of India's GDP and about 45% of exports [1], yet the Standing Committee on Finance (2022) estimated an MSME credit gap of ₹20–25 lakh crore, with less than 40% of units accessing formal credit [2]. Artificial Intelligence (AI) is now shifting underwriting from collateral-based to cash-flow-based lending, though its gains come with real risks.

How AI is reshaping MSME underwriting

• Alternate data over collateral: banks read GSTN filings, ITRs, bank statements and bureau scores instead of demanding physical security — SBI's in-house Business Rule Engine underwrote nearly ₹1 lakh crore of MSME loans (up to ₹5 crore each) in FY26 [3]. This is exactly the digital, cash-flow-based ecosystem the Standing Committee recommended [2].
• Speed and scale: sanction timelines have collapsed from weeks to minutes, easing working-capital stress and covering new-to-bank borrowers with no credit history [3].
• Better portfolio quality: SBI reported lower NPAs in the AI-underwritten book, while relationship managers were freed from manual data-gathering for higher-value work [3].
• Wider inclusion: thin-file micro units in semi-urban India gain a formal credit footprint, aiding financial inclusion.

Risks and regulatory challenges

• Opacity and bias: model-driven rejection is hard to explain; informal-sector borrowers with weak digital trails may be systematically excluded.
• Data privacy: pooling tax, GST and banking data heightens consent and purpose-limitation concerns under the DPDP Act.
• Accountability gap: grievance redressal for an algorithmic "no" remains undefined.
• Systemic risk: similar models across lenders can cause herding and correlated defaults; the RBI's FREE-AI Committee (2025), chaired by Dr. Pushpak Bhattacharyya, flagged bias, opacity and cyber vulnerability, offering seven "Sutras" and 26 recommendations on governance, audit and incident reporting [4].

AI is therefore a genuine enabler of credit democratisation, not a substitute for prudential judgment. Anchoring it in the FREE-AI framework — auditable models, human oversight and explainable rejections — can convert speed into trust, making the ₹20–25 lakh crore credit gap a closable one rather than a permanent feature.