Most fraud detection watches the transaction. This watches the conversation that leads to it.
Demo Flow ยท How It Works ยท Architecture ยท Setup ยท Privacy
Most UPI fraud isn't a hack โ it's a conversation. A caller impersonates a bank official, manufactures urgency, and walks the victim through authorizing their own transaction. From the bank's side, the payment looks completely legitimate: the real account holder approved it.
Common patterns:
| Pattern | Example |
|---|---|
| ๐ฆ Bank impersonation | "I'm calling from your bank's verification team." |
| โฐ Urgency + threats | "Your account will be blocked today." |
| ๐ OTP / PIN requests | "Share the OTP to confirm your identity." |
| ๐ Fake KYC claims | "Your KYC has expired, update it now." |
| ๐ธ Refund manipulation | "We're refunding โน5,000, just confirm the UPI PIN." |
| ๐ Investment scams | "Guaranteed 30% returns, act before market close." |
| ๐ฅ๏ธ Remote-access requests | "Install this app so I can help you." |
UPI Intent Guard reads the intent behind the payment โ not just the payment itself.
Three signal sources are fused into a single explainable risk score:
Conversation Context โ Scam Signal Detection โ Conversation Risk
Payment Details โ Intent Analysis โ Payment Intent Risk
Recipient Info โ Recipient Intelligence โ Recipient Risk
โ
โผ
Context Fusion Engine
โ
โผ
Final Risk Score (0โ100)
โ
โผ
Explainable User Intervention
| Score | Level | Response |
|---|---|---|
| 0โ30 | LOW | No action needed |
| 31โ60 | MEDIUM | Caution flagged |
| 61โ80 | HIGH | Strong warning shown |
| 81โ100 | CRITICAL | Payment intervention triggered |
"I'm calling from your bank." โ Bank impersonation
"Your account will be blocked today." โ Urgency + Account threat
"Transfer โน5,000 to verify your account." โ Verification request + Payment request
Output:
๐จ CRITICAL RISK โ 90/100
โ Bank impersonation
โ Urgency
โ Account threat
โ Verification request
โ Payment request
When a risky conversation is followed by a risky payment attempt, Context Fusion kicks in:
Conversation Risk 82
Payment Intent Risk 90
Recipient Risk 70
โโโโโโโโโโโโโโโโโโโโโโโโโโ
Final Risk 94 โ ๐จ CRITICAL
The user is shown a clear intervention screen โ "This payment may be part of a scam" โ with two explicit choices: STOP PAYMENT or CONTINUE ANYWAY. No real UPI transaction is ever performed by the prototype.
Scam-pattern detection works across English, Hindi, Telugu, and Hinglish โ because scam warnings only help if people can actually understand them in the language they think in.
Two demo modes are built in for reliable, repeatable presentation:
- Simulated Conversation โ six scripted scenarios (bank impersonation, OTP scam, KYC scam, refund scam, investment scam, and a safe control conversation) play out progressively for consistent hackathon demos.
- Microphone Demo โ an optional, controlled browser speech-recognition mode. This is not cellular call interception, and raw voice recordings are never stored.
| The prototype does not: |
|---|
| Perform real UPI transactions |
| Collect banking credentials |
| Store raw voice recordings |
| Require real OTPs |
| Access real bank accounts |
Core detection is designed to run locally, without requiring conversation audio to leave the device. Full details in SECURITY.md.
UPI INTENT GUARD
โ
โโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโ
โ โ
Conversation Payment
โ โ
โผ โผ
Signal Detection Intent Analysis
โ โ
โโโโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโ
โผ
Context Fusion
โ
โผ
Risk Engine
โ
โผ
Explainable Warning
โ
โผ
Intervention
|
Frontend
|
Backend
|
AI / Detection
|
Infra
|
UPI-intent-Guard/
โ
โโโ src/
โ โโโ components/
โ โโโ pages/
โ โโโ services/
โ โโโ models/
โ โโโ data/
โ โโโ hooks/
โ โโโ utils/
โ
โโโ backend/
โ โโโ app/
โ โโโ api/
โ โโโ models/
โ โโโ services/
โ โโโ ai/
โ โโโ risk/
โ โโโ tests/
โ
โโโ docs/
โโโ SECURITY.md
โโโ README.md
npm install
npm run devnpm test # run tests
npm run build # production buildcd backend
python -m pip install -r requirements.txt
python -m uvicorn app.main:app --host 0.0.0.0 --port 8000 --reloadpython -m pytest app/tests/ -v # run backend testsSTART DEMO
โ
โผ
Simulated Bank Call
โ
โผ
"Your KYC has expired." โ Risk โ
โ
โผ
"Your account will be blocked." โ Risk โ
โ
โผ
"Transfer โน5,000 to verify." โ ๐จ CRITICAL RISK
โ
โผ
Payment Simulator โ Context Fusion โ 94/100
โ
โผ
STOP PAYMENT
โ
โผ
๐ก๏ธ โน5,000 PROTECTED
All demonstrations run on synthetic data. No real money ever moves.
- Android implementation with deeper device-level integration
- Optimized on-device multilingual speech models
- Improved scam-intent classification
- Stronger recipient intelligence
- Accessibility features for elderly users
- Integration with legitimate payment apps (where platform policy permits)
UPI Intent Guard is a hackathon prototype for demonstrating proactive, conversation-aware scam detection. Risk scores are experimental and should not be treated as a guarantee against fraud. The current web MVP does not intercept cellular calls or perform real financial transactions.
Built to protect the moment before the payment โ not just the payment itself.