Jennie AI

Verify the story. Not just the identity.

Inspiration

What if the biggest weakness in banking security isn't a stolen password, but a legitimate customer being manipulated into sending money?

Traditional fraud detection systems focus on unauthorized transactions, suspicious logins, and stolen credentials. But modern scammers have found another way: convincing people to willingly authorize fraudulent payments.

A victim might pass every security check and still lose their savings.

That inspired Jennie AI, an AI-powered scam intervention system designed to identify manipulation behind authorized payments.

What It Does

Jennie AI goes beyond traditional fraud detection by asking a different question:

"Is this transaction authorized, or is someone manipulating the person authorizing it?"

The system is designed around five capabilities:

1. Intelligent Risk Detection: Identifies potentially suspicious transactions using unusual payment amounts, new recipients, transaction history, and contextual warning signs.

2. AI Verification Missions: Instead of displaying generic fraud warnings, Jennie AI creates personalized verification steps based on the member's situation.

3. Independent Verification: Helps members verify payment requests using trusted contact information rather than phone numbers, links, or instructions supplied by potential scammers.

4. Smart Intervention: Gives members options to pause an eligible payment, cancel it, request assistance, or continue through the appropriate review process.

5. Post-Transaction Protection: Explores how suspicious activity discovered after payment authorization can trigger rapid reporting, recovery guidance, and employee escalation.

How We Built It

it is designed as a member-facing scam prevention experience with an AI-assisted risk assessment and verification workflow.

The architecture separates transaction risk detection, contextual scam analysis, verification recommendations, and member decision-making.

The system emphasizes explainable warnings, straightforward user interactions, and human oversight for sensitive financial decisions.

Challenges We Faced

One of the biggest challenges is distinguishing a legitimate unusual payment from a scam.

Blocking every suspicious transaction would frustrate customers, while allowing every authorized payment could expose members to manipulation.

Another challenge is designing interventions that protect members without taking away their ability to make legitimate financial decisions.

We also explored the challenge of what happens after verification, especially when a member proceeds with a transaction or discovers a scam too late.

What We Learned

We learned that verifying someone's identity is not the same as verifying the legitimacy of their transaction.

Effective scam prevention requires understanding behavioral manipulation, reducing unnecessary friction, and giving people practical ways to verify suspicious requests.

What's Next

Our next development goals include:

• Improved detection of manipulation patterns.
• More personalized verification missions.
• Post-transaction scam reporting and recovery workflows.
• Employee escalation with explainable risk summaries.
• Better evaluation of false positives and legitimate payment experiences.

Our Vision

We want to move financial security beyond simply asking whether the right person authorized a transaction.

We want to help people recognize when they are being manipulated before their money is gone.

Jennie AI — Verify the story. Not just the identity.

Built With

• ai
• fastapi
• javascript
• openstreetmap
• python
• react