Startup Originals Team
To combat rising digital frauds, the Reserve Bank of India (RBI) is building an AI-powered system that can detect and alert users about risky online transactions in real time. The system, called the Digital Payments Intelligence Platform (DPIP), is being prototyped currently by the RBI’s innovation arm and is expected to enhance fraud prevention across India’s digital payments ecosystem.
What Is the Digital Payments Intelligence Platform (DPIP)?
- DPIP is an AI system under development that will collect and analyze data from multiple sources, such as banking records, telecom data, geographic information, and known mule accounts, to flag suspicious transactions.
- If a transaction is deemed “risky,” DPIP will generate a pre-transaction alert, giving banks or users the option to halt or proceed with the payment.
- The platform will be run by a separate entity (distinct from existing banking systems) to maintain neutrality and efficacy.
Why RBI Is Doing This:
- India has seen significant losses from cyber fraud: in the first nine months of FY25 alone, ₹107.21 crore was lost due to cyber frauds.
- In FY25, the number of digital frauds involving cards and internet banking totaled 13,516 cases, amounting to ₹520 crore across India.
- With more people relying on digital payments (UPI, wallets, QR scanning), the RBI sees advanced detection systems as essential to sustaining trust in the system.
How It Will Work & Key Features:
| Feature | Description |
|---|---|
| Data Sources | Banking data, telecom info, location data, mule account info, etc. |
| Alert Mode | Generate real-time alerts before a transaction completes |
| User Choice | The system will notify but leave the decision with the user or bank whether to proceed |
| New Entity | The platform will be managed by a separate body to isolate it from day-to-day banking operations |
Challenges & What to Watch:
- Privacy and data sharing: Aggregating banking, telecom, and location data raises questions around user privacy, data rights, and regulatory compliance.
- False positives: Overly aggressive flagging could inconvenience legitimate users by blocking genuine transactions.
- Scalability & accuracy: The model must balance speed and precision as it scales across India’s massive transaction volume.
- Adoption and integration: Banks, fintechs, and payment providers need to adapt to alerts and incorporate workflows to respond effectively.
Conclusion:
RBI’s AI initiative to flag risky online payments is a bold and necessary step toward strengthening India’s defense against digital fraud. If implemented well, with care for privacy, handling scale, and maintaining low false alerts, it could transform how financial fraud is detected and prevented in the country.
