Piramal Finance Limited · NSE:PIRAMALFIN
Piramal Finance — Investec AI Unmasked - The Real & Possible Conference 2025
59 slides from Piramal Finance's conference presentation, Dec 12, 2025
Slide 1: Intimation of Investor Presentation to Stock Exchanges
12th December, 2025 BSE Limited National Stock Exchange of India Limited 1st Floor, New Trading Wing, Exchange Plaza, 5th Floor, Rotunda Building, P. J. Towers, Plot No. C/1, G Block, Dalal Street, Fort, Bandra-Kurla Complex, Bandra (E), Mumbai - 400 001 Mumbai - 400 051 BSE Scrip Code: 544597 NSE S
Slide 2: Piramal.ai: Inside the core of AI transformation at Piramal
6 Piramal Finance Piramal.ai Inside the core of Al transformation at Piramal
Slide 3: Table of contents
A Piramal Finance – Our Journey So Far 03 B Tech, Data Science & AI - Our Approach 18 Table of contents C The Evolution Of AI 27 D AI In Action @ Piramal Finance 40 E Recognitions for AI Excellence 49 F What This Means For You As Investors 55
Slide 4: Piramal Finance – Our Journey So Far
Piramal Finance – Our Journey So Far 3
Slide 5: We are now a retail-led NBFC. We have grown ~4X in 4 years
We are now a retail-led NBFC. We have grown ~4X in 4 years. (post merger with DHFL) Our AUM mix has fully pivoted; new Growth book is now 94% of total AUM; ₹74,704 Cr Retail AUM In ₹ k Cr Growth (Retail + WS2.0) 91 Legacy 6% 5 81 11 7 94% 69 65 64 9 15 Total AUM 49 6 29 Legacy 43 WS2.0 75 3 Retail 6
Slide 6: Our retail business has scaled faster than any peer NBFC
Our retail business has scaled faster than any peer NBFC Indexed AUM growth Time taken for Retail NBFCs* AUM to grow from ₹20k Cr to ₹75k Cr Rebasing Sep-21 at 100 345 Retail AUM ₹ k Cr Piramal Retail 325 Piramal Consol. 75 305 Retail Peers* 285 265 65 245 225 55 205 185 45 165 DHFL acquisition 145
Slide 7: What is driving this level of Growth?
What is driving this level of Growth? 6
Slide 8: Growth Driver #1: Product expansion
Growth Driver #1: Product expansion We have been steadily expanding our product portfolio 19 UBL Gold Building a multi-product strategy across Co-branded credit card less-penetrated banking segments Microfinance-Organic # Total No. of Products Loans Against Shares 13LAP Micro Additions during the ye
Slide 9: Growth Driver #2: Geography expansion
Growth Driver #2: Geography expansion We have built a distribution infrastructure of 517 branches across all parts of the country Pan India presence Scale up of branch network No. of branches Share in retail AUM % 1 518 518 76 Conventional MFI Branches 1 Branches 19 1 8 301 19 13 9% 40 46 11% 6 6 14
Slide 10: Growth Driver #3: Cross Sell
Growth Driver #3: Cross Sell 11% of all disbursements (27% of unsecured) now come through cross-sell Sep 2025 Total customer franchise at 5.2 Mn In Mn Cross-sell% in unsecured disbursements Total customer franchise 5.2 11,499 11,099 7,133 7% 23% 27% Credit segment filtered customers 3.9 Unsecured Di
Slide 11: Growth Driver #4: Productivity
Growth Driver #4: Productivity We have seen significant productivity gains, driving growth as well as operating leverage 1 AUM per branch In ₹ Cr 2 Revenue per branch Opex ratio reduced ~260 bps over last 10 quarters Indexed at 100 as on Mar-23 144 Trailing 12 months 179 6.75% 6.5% 6.50% 1.8x 6.25%
Slide 12: We were early movers in tech–fintech partnerships
We were early movers in tech–fintech partnerships, and remain significant players here Embedded Finance overview Fintech Partners we have worked with.. Piramal Finance stats on Fintech partnerships since Dec ’20: More than 40 programs launched with 25+ partners Personal/consumption loans 15,000+ 28,
Slide 13: What about Credit Risk outcomes?
What about Credit Risk outcomes? 12
Slide 14: We have been able to maintain a stable risk profile in a volatile environment
We have been able to maintain a stable risk profile in a volatile environment Net 90+ (90 to 180 DPD) Horizontal Risk – 6m30+ and 12m90+ 6m30+ 12m90+ 3.7% 2.8% 2.3% 2.2% 2.6% 2.1% 2.1% 2.4% 1.9% 1.8% 2.2% 1.6% 1.6% 1.4% 1.8% 1.1% 1.7% 0.7% 0.7% 0.7% 0.8% 0.8% 0.8% 0.8% 1.4% 1.4% 0.7% 0.6% 0.7% 0.6%
Slide 15: Our multi-product strategy enables portfolio-level stability
Our multi-product strategy enables portfolio-level stability despite product-level volatility Housing loans 29,550 32% LAP 21,285 23% Used car loans 4,797 5% 3% 3% 3% 2% 2% 1.6% 1.5% 2% 1.3% 1.3% 1.1% 1.0% 1.0% 1.1% 0.9% 0.9% 1.0% 0.9% 0.8% 0.9% 0.8% 0.7% 0.7% 0.6% 0.6% 0.6% 0.7% 0.7% 0.6% 0.6% 1% 0
Slide 16: New origination quality has remained stable as disbursements ramped up
New origination quality has remained stable even as disbursement volumes have ramped up Housing loans 29,550 32% LAP 21,285 23% Used car loans 4,797 5% 6.3% 6% 6% 6% 5% 5% 3.9% 5% 4% 4% 2.6% 2.4% 2.3% 2.7% 2.3% 2.5% 3% 3% 2.3% 2.3% 3% 1.5% 2% 2% 0.9% 0.6% 0.4% 0.4% 0.5% 0.4% 0.5% 0.3% 0.5% 0.5% 0.6%
Slide 17: Profitability has stabilised and started a steady ascendence
After a tumultuous period in FY23-24, profitability has stabilised and started a steady ascendence Volatility in net profit is reducing Consol. PAT, in ₹ Cr. Growth business PBT In ₹ Cr. 1,300-1,500 8,155 3,545 509 327 276 ~₹250 Cr impact on 181 163 profits due to retail 137 fee amortization 102 48
Slide 18: We are guided by a strong & experienced leadership team
We are guided by a strong & experienced leadership team driving the scale-up Jairam Sridharan MD & CEO 25+ years 25+ years 20+ years 20+ years 15+ years 20+ years 20+ years Jagdeep Mallareddy Sunit Madan Saurabh Mittal Markandey Upadhyay Arvind Iyer Nirav Adani Parneet Soni CEO – Retail Chief Operat
Slide 19: Tech, Data Science & AI: Our Approach
Tech, Data Science & AI: Our Approach 18
Slide 20: Four foundational choices on Tech
At the core of our AI capabilities are four foundational choices on Tech In-house engineering talent Agility | Tech Products | UX 1 first | Innovation and systems Quality engineering 2 and DevOps Tech Product Culture 3 Distributed computing Architectural choices | Innovation | Cost efficiency | Scal
Slide 21: Our analytics journey stands at the cusp of an Agentic universe
On the analytic side, our journey started modestly, and stands today at the cusp of an Agentic universe Agentic AI “Building Autonomous AI-driven agents” Deep Learning “Unstructured Data Machine Learning Intelligence” (Predictive AI) “Predict customer outcomes” Decision Science “Build Prescriptive A
Slide 22: The right talent model for AI has to fuse both – Tech & Math
The right talent model for AI has to fuse both – Tech & Math AI at Piramal Finance – key actors driving transformation Saurabh Mittal Markandey Upadhyay Chief Technology Officer Chief Data & Analytics Officer Jaydeep Chakrabarty Karan Arora Vasuki Setlur Tushar Satish Kaushik Deb Ravi Shankar AI in
Slide 23: We have staffed entirely novel roles for scalable AI
… and we have staffed entirely novel roles for scalable AI AI Product Management AI Manufacturing AI Deployment AI Consumption and Governance AIPM Tech DPMs Agentic Manufacturing Agentic Operations (AI Product Managers) Digital Product Managers Foundation Models & AI Annotation & Principal Architect
Slide 24: Our AI, Tech & data science capabilities have achieved massive scale
Our AI, Tech & data science capabilities have achieved massive scale 1 ~70,000 Sq ft 2 300+ 3 200+ Space Bengaluru Team of product engineering AI/Data scientists Centre of Excellence 4 10,000+ 5 350+ 6 150 Features analysed for Internal dashboards AI/ML/Agentic models* underwriting 7 45+ 8 30+ 9 120
Slide 25: A culture of learning, experimentation, and campus engagement
The fuel powering our AI efforts is a culture of learning, experimentation, and campus engagement We invest in people & skills via training, conferences, hiring We run Hackathons with top IITs, BITs to innovate on new technologies 9 Candidate Screening basis GEN AI Workshop At CYPHER Conference Appl
Slide 26: Which has brought us to Stage 3 of Tech transformation
Which has brought us to Stage 3 of Tech transformation … Stage - 1 Stage - 2 Stage - 3 Digital Transformation Cloud Native Transformation AI Native Transformation Digitized our processes Enabled agility and scale for the Intelligence & Productivity @ scale business 25
Slide 27: Enabled us to build a truly AI-Native ecosystem
…and enabled us to build a truly AI-Native ecosystem In the Business Piramal’s AI-Native Ecosystem 26
Slide 28: The Evolution Of AI
The Evolution Of AI 27
Slide 29: From Analyst to Partner: The 3 great stages of AI
From Analyst to Partner: The 3 great stages of AI AI is not one technology, but three distinct stages of capability: from AI that analyses, to AI that advises, and finally, to AI that acts. 1 2 3 Predictive AI Generative AI Agentic AI The Analyst The Advisor The Assistant “What will happen?” “What s
Slide 30: Predictive AI: Given a feature set, predict a class
Stage 1 The challenge Given a feature set, predict a class 29
Slide 31: Predictive AI: What is this?
What Model predicts – It’s a cat is this? Predictive AI Given some features, AI predicts a class Image by Images leonardo.ai 30 by leonardo.ai
Slide 32: Generative AI: Given a class, generate something novel
Stage 2 The challenge Given a class, generate something novel 31
Slide 33: Generative AI
Generative AI Given a class, AI generates content on demand, with some noise Model creates a cat (“creative freedom”) formally dressed I want a cat dressed formally Images by leonardo.ai 32
Slide 34: Agentic AI: Given a problem, assist me with a solution
Stage 3 The challenge Given a problem, assist me with a solution 33
Slide 35: Book my cat's Saturday appointment with a top-rated vet
Book my cat’s Saturday appointment with a top- rated vet and add it to my calendar 34
Slide 36: Agentic AI
Agents autonomously found the top vet, scheduled the appointment, Agentic AI and set reminders. Researcher Agent Action Agent Organizer Agent Finds options via Navigates the Creates the Web Search & clinic's website to calendar event on Maps and secure the user’s calendar analyses ratings. appointme
Slide 37: Agentic AI
Agents autonomously found the top vet, scheduled the appointment, Agentic AI and set reminders. Given a goal, AI agents execute tasks to achieve an outcome.
Slide 38: The AI space has been evolving rapidly since Nov 22
The AI space has been evolving rapidly since Nov 22. At Piramal, we have stayed right at the leading edge. Industry Launch Nov 22 Nov 24 Nov 24 Apr 25 Apr 25 Production Aug 23 Apr 25 Apr 25 Jul 25 Aug 25 @Piramal LLM MCP Agents A2A Agentic Ecosystem Foundational model Model context protocol AI Agent
Slide 39: "95% of AI pilots are failing" – MIT; "We are in the 5%"
“95% of AI pilots are failing” – MIT “We are in the 5%” - Piramal 95% of generative AI pilots Small Language Models are How do we stay at companies are failing the Future of Agentic AI in the 5%? ▪ Misaligned with Operations ▪ Small language models (SLMs) ▪ Business-First ▪ Most fail due to brittle
Slide 40: Purpose-built, business-aligned AI, not generic LLMs, for lending
We are focused on purpose-built, business-aligned AI, not generic LLMs, for lending Hybrid AI Engine Open Culture of Architecture Experimentation Marry Predictive and Business Generative Al. Flexible 'plumbing' that Strong Talent Sponsorship allows all systems to share R&D mindset data instantly. Co
Slide 41: AI in Action @ Piramal Finance
AI in Action @ Piramal Finance 40
Slide 42: Piramal.ai 45+ live use cases across the customer lifecycle
Piramal.ai 45+ live use cases across the customer lifecycle Fulfillment ❑ AI Agent driven Field Investigation Underwriting Post Disbursement ❑ Automated docket verification ❑ Bank Statement Intelligence ❑ AI Customer onboarding ❑ Sanctioned Undisbursed Conversion ❑ Fraud Assessment ❑ Automated email
Slide 43: Illustrative use cases of Piramal.ai
Let us look at a few illustrative use cases of Piramal.ai 1 ARYA* In house AI everyday assistant, coach & work friend 2 AALOK Bank statement intelligence 3 LEO + PRISM Fraud decisioning engine & tampered document detection tool 4 CAM COPILOT Credit Assessment Memo 5 DOC.AI Document intelligence 6 NI
Slide 44: Top 5 Outcomes of Piramal.ai
1 Top 5 Outcomes of 5 Integrate & Underwrite Build More Piramal.ai Better 2 4 3 Enhance Improve Productivity CX Drive Growth 43
Slide 45: Underwrite Better
Horizontal Risk (30+ at 6MOB) down YoY by Fraud Risk (Straight flow 90+) down YoY by 40%+ for each of the 3 consecutive years >50% for each of the 3 consecutive years 5.6% 4.7% 4.7% 3.7% 2.9% Underwrite 1 Better 2.3% 2.4% 2.1% 1.8% 1.8% 1.7% 1.5% 1.3% 1.3% 1.1% 1.1% 1.0% 0.8% 0.6% 0.6% 0.4%0.4% 0.3%
Slide 46: Enhance Productivity
Productivity up YoY across Lending Functions Sales Productivity Credit Manager Productivity 28 68 60 22 Enhance 2 H1 FY25 H1 FY26 H1 FY25 H1 FY26 Productivity Overall Monthly Disb. Productivity (₹ lakh / month / Files Processed per Credit Manager (Rolling 12 month) salesperson) for branch-based reta
Slide 47: Drive Growth: Consistent disbursement growth over years
Consistent disbursement growth over years FY23 FY24 FY25 FY26 10,954 9,892 8,718 8,362 Drive 3 8,065 8,910 Growth 6,816 7,692 6,828 6,246 5,707 5,111 3,973 2,459 Q1 Q2 Q3 Q4 46
Slide 48: Improve CX
Quarterly Complaints (per 1000 customers) down by 85% 2.8 2.8 2.6 2.5 2.1 1.8 Improve 4 CX 1.2 0.4 0.3 0.4 0.4 0.3 Q3FY23 Q4FY23 Q1FY24 Q2FY24 Q3FY24 Q4FY24 Q1FY25 Q2FY25 Q3FY25 Q4FY25 Q1FY26 Q2FY26 47
Slide 49: Build More
Build More 5 48% of overall code now written by AI 48
Slide 50: Recognitions for AI Excellence
Recognitions for AI Excellence 49
Slide 51: Google Agentic AI Day runners-up and Gartner finalists
2nd Runners-up at Google Agentic AI Day (9,000+ teams competed) Finalists at Gartner’s 2025 Eye on Innovation Awards for Banking & Investing, APAC (Final results yet to be announced) 50
Slide 52: Best NBFC of the Year for AI-powered CX
Best NBFC of the Year for AI-powered CX at 10th NBFC Summit & Awards 2025 Best AI Chatbot Product of the Year, at The Brainalytics: Arya 51
Slide 53: Top 50 Tech Leaders in NBFC
Top 50 Tech Leaders in NBFC - ET NBFC Leadership Conclave 2025 Global Tech Influencer of the Year - Emirates Group, Dubai Nov 2024 Author: Author: The Hindu column “The Value Vector” – Tech and Context 52
Slide 54: We have filed patents for our underlying AI solutions architecture
We have filed patents for our underlying AI solutions architecture 53
Slide 55: We continue to expand our engagement with the partner ecosystem
We continue to expand our engagement with the partner ecosystem across the spectrum Technology Partners Product Vendors Invited senior leadership team to MS We use ~8 Bn tokens per month with headquarters where they will showcase Part of their Red team. Helping testing Claude only to write code. the
Slide 56: What This Means for You As Investors
What This Means for You As Investors 55
Slide 57: Piramal.ai strategy is leading to tangible benefits
4 Piramal.ai strategy is leading to tangible benefits across several dimensions 1. Underwrite Better 2. Enhance Productivity 3. Drive Growth Vintage Risk (30+ at 6MOB) - retail Productivity up YoY across Retail Lending Functions Consistent retail disbursement growth In ₹ Cr. FY23 FY24 FY25 FY26 5.6%
Slide 58: Additional 25bps Opex savings and a stronger AUM growth outlook
With proven AI adoption success, we are baking in additional 25bps Opex savings and a stronger AUM growth outlook Revising Retail Opex-AUM outlook lower We expect AUM growth momentum to sustain longer In ₹ L Cr 20-25% 2.0+ 6.75% CAGR 6.5% 6.50% 6.25% 6.00% 5.8% 5.8% 1.5+ 5.75% 5.5% 5.50% 5.3% 5.25%
Slide 59: Thank You
\ Thank You
Piramal Finance — Investec AI Unmasked - The Real & Possible Conference 2025 on 6ix