Inspiration
Cross-border payments can be confusing, especially for people who are not familiar with blockchain, Web3, transaction fees, network speeds, and security risks. A user may know exactly what they want to do — for example, "I want to send $500 from the US to India" — but understanding the best route, estimated cost, settlement time, and potential risks can be difficult.
We wanted to make Web3 payments feel less complicated and more understandable.
This inspired us to build TrustPay AI, an AI-powered payment assistant that turns a simple natural-language request into an understandable payment plan. Instead of forcing users to understand the technical details first, TrustPay AI explains those details to the user and helps them make a more informed decision.
Our goal was to combine AI + Web3 + explainability into one practical application rather than building another generic chatbot.
What it does
TrustPay AI is a Web3 payment assistant designed to help users understand and simulate cross-border transactions.
A user can describe what they want to do in natural language, such as:
"I want to send $500 from the US to India."
TrustPay AI processes the request and provides:
- AI-powered payment understanding — extracts important information such as amount, currencies, destination, and user preferences.
- Payment route comparison — compares simulated routes based on cost, speed, risk, and transparency.
- AI route recommendation — recommends a route based on the user's priorities, such as lowest cost, fastest settlement, lower risk, or a balanced option.
- Explainable recommendations — users can see why a particular route was recommended instead of receiving a black-box decision.
- Risk analysis — evaluates simulated transaction characteristics and provides a Low, Medium, or High risk assessment with understandable reasons.
- Web3 transaction simulation — demonstrates what a blockchain payment flow could look like, including a simulated transaction hash and confirmation.
- Transaction history — allows users to review their previous simulated transactions.
- Demo scenarios — provides ready-to-use scenarios so judges and users can experience the complete product without configuring real financial accounts.
All transactions are simulated. TrustPay AI does not process real money or collect banking/card information.
How we built it
We designed TrustPay AI as a modular web application so that the AI assistant, recommendation system, risk analysis, and Web3 simulation could work together while remaining easy to understand.
The main components are:
AI Payment Assistant
The assistant accepts natural-language payment requests and converts them into structured information.
For example:
Input: "Send $500 from the US to India as cheaply as possible."
Extracted information:
- Amount: $500
- Source: US
- Destination: India
- Preference: Lowest cost
The structured request is then passed to the recommendation system.
Route Recommendation Engine
We created a scoring system that evaluates simulated payment routes using factors such as:
- Transaction fee
- Settlement time
- Risk score
- Transparency
- User preference
The system then produces a recommendation and explains the reasoning behind it.
This allows the same payment request to produce different recommendations depending on whether the user prioritizes cost, speed, safety, or balance.
Risk Analysis
TrustPay AI includes a simulated risk-analysis layer that looks for unusual characteristics in a transaction.
Examples include:
- Unusually large transactions
- Unusual destinations
- Repeated transaction patterns
- Inconsistent recipient information
Instead of simply displaying a risk score, the system explains the factors contributing to the result.
Web3 Simulation
To demonstrate the blockchain experience without handling real cryptocurrency, we created a simulated transaction system.
When a user confirms a demo payment, the application generates a simulated transaction hash and walks through a transaction lifecycle such as:
Preparing → Processing → Confirming → Confirmed
This gives users a realistic understanding of how a Web3 transaction could appear without exposing them to real financial risk.
User Interface
We focused on creating a clean fintech-style interface rather than making TrustPay AI look like a basic chatbot.
The interface includes:
- AI payment assistant
- Route comparison cards
- Risk indicators
- Explainability panels
- Transaction statistics
- Transaction history
- Interactive demo scenarios
- Responsive dashboard
Challenges we ran into
One of our biggest challenges was finding the right balance between technical complexity and usability.
Web3 contains many concepts that can be difficult for beginners to understand. Simply displaying technical information would not solve the problem we were trying to address.
We therefore focused on translating complex information into simple explanations.
Another challenge was designing a recommendation system that did not simply choose the cheapest route. A cheap route may be slower or have a higher simulated risk. We needed to consider multiple factors simultaneously.
We also had to make the project useful as a hackathon prototype without connecting it to real financial infrastructure. We solved this by building a realistic simulation environment where users can experience the complete workflow without real money or sensitive financial information.
Finally, making the entire experience understandable in a short demonstration was important. We designed a guided demo flow that allows someone to enter a payment request and see the complete journey from AI interpretation to route recommendation, risk analysis, and simulated blockchain confirmation.
Accomplishments that we're proud of
We are proud that TrustPay AI brings several technologies together into one coherent experience.
🤖 AI + Web3
Instead of treating AI and Web3 as separate features, we connected them around a real-world problem: making cross-border payments easier to understand.
💡 Explainable AI
A major part of TrustPay AI is not just giving users an answer, but explaining why the system reached that recommendation.
🔐 Safer experimentation
The complete payment experience can be demonstrated without using real money, bank accounts, or cryptocurrency.
📊 Multi-factor recommendations
The recommendation system considers several factors instead of optimizing for only one metric.
🎯 Real-world usability
The project is designed around a problem that ordinary users can understand immediately, rather than being built purely as a technical demonstration.
What we learned
Building TrustPay AI taught us that a good AI application is not necessarily the one with the most complicated AI model.
The most important part is how effectively the technology solves a problem for the user.
We learned how to:
- Turn natural-language requests into structured data.
- Design explainable AI recommendations.
- Build multi-factor scoring systems.
- Create realistic Web3 transaction simulations.
- Design interfaces for technically complex systems.
- Think about security and privacy when designing financial applications.
- Balance technical ambition with a reliable hackathon prototype.
- Build a product around a complete user journey rather than isolated features.
We also learned that trust is just as important as intelligence in financial technology. If an AI recommends something, users should be able to understand the reasoning behind that recommendation.
What's next for TrustPay AI
TrustPay AI is currently designed as a prototype and simulation environment. Our next goal would be to move from simulated infrastructure toward carefully designed integrations with real Web3 networks and payment infrastructure.
Future versions could include:
- Real-time network fee estimation
- Support for additional blockchain networks
- Real-time exchange-rate information
- Wallet integration
- On-chain transaction verification
- More advanced anomaly detection
- Personalized payment preferences
- Multi-language support
- Improved AI financial explanations
- Greater transparency around transaction costs
- Real-time transaction tracking
However, safety and user control would remain central to the product.
Our long-term vision is to make Web3 payments feel less intimidating by putting an intelligent, transparent layer between users and complex financial infrastructure.
TrustPay AI is built around a simple idea: users shouldn't need to understand the complexity behind a payment in order to understand the decision they're making.
Built With
- next.js
