PhonePe is using AI to handle customer support at scale, helping its small support team manage billions of transactions and millions of users.
PhonePe AI is helping the fintech company manage 10 billion monthly transactions, automate customer support, and improve operations without dramatically expanding its workforce.
At the AI Tech Talk during the Global Fintech Fest 2026, Rahul Chari, founder and CPTO of PhonePe, explained how the company uses artificial intelligence across its consumer products, merchant services and internal operations.
The underlying approach is straightforward. Remove friction wherever users encounter it.
For consumers, that can mean finding a service without navigating multiple menus. For merchants, it can mean getting a business online without hiring a developer. For engineers, it can mean diagnosing crashes or analysing threats in minutes rather than days.
PhonePe has built AI systems around each of those problems.
PhonePe Uses AI to Understand User Intent
Finding a specific service inside a financial app can become difficult when users have to move through several layers of menus.
PhonePe’s AI Search aims to remove that problem.
Users can type or speak what they want, such as buying gold, and the system interprets the request and takes them toward the relevant transaction screen. The system can also use information already available to pre-fill details, reducing the number of steps required to complete an action.
PhonePe applies a similar approach to personal financial information.
Its AI Transaction Insights allows users to ask questions about their spending patterns. A request about household expenses, for example, can generate a visual breakdown across categories such as electricity, gas, mobile services, and broadband.
The system can then provide suggestions around saving and investing based on the spending information available within the app.
Agentic AI Simplifies Insurance
Insurance presents another problem. Customers often face complicated terminology, multiple plans, and too many choices.
PhonePe is using conversational AI to simplify that process.
A customer looking for car insurance can enter the vehicle’s registration number. The AI agent can then identify relevant plans and guide the customer through the selection process.
Once the customer chooses a policy, the agent continues to guide the process through to issuance.
The goal is not simply to add an AI chatbot. PhonePe is using AI to reduce the number of decisions and steps a customer has to navigate manually.
AI Helps Small Businesses Go Online
PhonePe is also applying AI to its merchant ecosystem.
Small businesses selling flowers, handicrafts, homemade food and other products may not have access to developers or web designers. PhonePe’s AI-powered PG SmartPages allows merchants to describe their products, including in vernacular languages.
The system then generates product copy, selects relevant fields and creates a mobile-optimised, payment-ready page in under 10 minutes.
That can reduce a process that traditionally requires technical assistance to something a small business owner can complete independently.
PhonePe Cuts Payment Integration From Weeks to Minutes
Larger digital businesses face a different problem when integrating payment infrastructure.
Traditional payment gateway integration can require weeks of engineering work. PhonePe’s AI Integration Agent aims to shorten that process.
The agent works with PhonePe PG’s existing API specifications, helps correct payloads, and supports error handling through a conversational workflow.
By reducing integration time, PhonePe aims to free engineers to spend more time improving their products rather than resolving repetitive implementation issues.
PhonePe Opens Market Data Through AI
PhonePe is also using AI to make its market data easier to access.
PhonePe Pulse provides market intelligence that businesses and entrepreneurs can use to understand activity at a more local level. An entrepreneur planning a business in Punjab, for example, can use the available data to examine market trends and make decisions based on more recent information.
At the Global Fintech Fest session, Chari introduced Ask Pulse, an AI-powered chatbot that lets users search for PhonePe Pulse insights through natural-language conversations.
The move turns a large data resource into a more accessible research tool.
AI Is Changing PhonePe’s Engineering Operations
The consumer-facing products sit on top of an AI-enabled engineering operation.
PhonePe has developed internal tools to reduce the time engineers spend on debugging, threat analysis, and repetitive coding tasks.
TraceMind, an internal debugging tool, helps engineers investigate crashes and analyse threats. PhonePe says the system can reduce work that previously took days to minutes while supporting the stability of its core platform.
Another internal tool, Lens, provides real-time assistance for UI generation and mobile debugging.
Then there is KadhAI, which focuses on code orchestration.
The system can plan and implement code changes inside isolated sandboxes, run tests and checks, and send the work for human approval. PhonePe says the system has reduced code scaffolding time from two to three hours to about 15 minutes.
That allows engineers to spend more time on system design and less time on repetitive coding work.
More Than 200 AI Agents Run Inside PhonePe
PhonePe has extended its use of AI beyond engineering.
Its internal Agent Hub provides a platform for deploying AI agents securely across engineering, business, and corporate functions.
The platform now supports more than 200 internal agents. Together, they handle more than 1,600 queries each day for 759 weekly active users.
PhonePe says the automation saves roughly five minutes per task. In some departments, AI handles as much as 50% of daily tasks.
These numbers show where the company sees one of the biggest opportunities for AI. Not simply replacing individual tasks, but reducing the amount of repetitive work employees need to handle.
PhonePe Automates More Than 92% of Customer Support Tickets
The scale becomes clearer in customer support.
PhonePe processes around 10 billion transactions every month with about 500 support agents.
The company says AI now automates more than 92% of customer support tickets. At that scale, automation allows a relatively small support operation to serve a massive transaction base.
PhonePe says its AI-driven support operation has generated 40 times more value while keeping the team lean.
Fraud investigations show a similar shift.
AI now handles around 80% of end-to-end fraud investigations, including complex cases involving multiple forms of information. Processes that previously took around 90 minutes can now be resolved almost instantly.
The cost per investigation has also fallen from roughly ₹50 to between ₹5 and ₹15.
PhonePe’s AI Strategy Is Built Around Scale
The common thread across PhonePe’s AI initiatives is not a single product.
It is scale.
A voice command that takes a customer directly to a service. An AI tool that puts a small merchant online in minutes. An agent that reduces payment integration work. An internal system that helps engineers debug code. An automated fraud investigation that costs a fraction of the previous process.
Each use case addresses a different problem, but the objective remains the same.
Reduce friction, automate repetitive work,k and reserve human effort for decisions that require judgment.
PhonePe’s AI strategy therefore goes beyond adding artificial intelligence to an existing fintech product. The company is embedding AI across the customer journey, merchant ecosystem, engineering stack, and internal operations.
At 10 billion transactions a month, even small improvements can create enormous operational effects. That is where PhonePe’s approach to AI is becoming most significant.
