TrustWise AI

TrustWise AI is a self-reported full-stack escrow dispute resolution platform that combines blockchain, cryptographic verification (via TLSNotary), and large language models (GPT-4o) to automate trust in online commerce. The system allows users to create escrows on Ethereum Sepolia, submit disputes with minimal evidence (tracking number, buyer/seller claims), and receive AI-generated arbitration decisions based on cryptographically verified shipment data.

The platform is built as a React frontend with Node.js/Express backend, integrating Solidity smart contracts, Rust-based TLSNotary verification, and GitHub Models for GPT-4o. It aims to replace traditional screenshot-based dispute resolution with a system where AI decisions are grounded in authenticated evidence.

Key claims include:

• AI arbitration decisions are based on cryptographically verified evidence
• No KYC required
• Smart contracts hold funds
• Zero-trust architecture using TLSNotary proofs

The description states this is a hackathon project built by two team members, submitted to the OpenAI 2026 hackathon. There is no evidence of revenue, customers, or traction beyond the authors' own account.

The description states TrustWise AI is a "full-stack escrow dispute resolution platform that combines blockchain, AI, and cryptographic verification into a single workflow."

Based on the author's own write-up:

• It creates escrows using Ethereum smart contracts deployed on Sepolia
• Users submit disputes by entering tracking number, buyer claim, seller claim
• The system automatically fetches shipment data and generates TLSNotary proofs
• GPT-4o (via GitHub Models) analyzes verified evidence to generate structured arbitration decisions
• Settlement occurs through MetaMask on-chain

The system architecture shows:

• Frontend: React/Vite dashboard with wallet integration (MetaMask)
• Backend: Node.js/Express API orchestrating blockchain, AI, and verification
• Smart contracts: Solidity on Ethereum Sepolia
• Verification layer: Rust-based TLSNotary verifier
• AI engine: GPT-4o through GitHub Models

The product is described as a complete end-to-end system for escrow dispute resolution that integrates these technologies.

The description states the platform's core positioning:

• "Decentralized escrow combining GPT-5.6 AI arbitration with zkTLS web proof instead of screenshots"
• "Smart contracts hold funds; AI instantly resolves disputes privately with zero KYC required"

The authors claim this addresses a fundamental problem in online commerce: "Trust is one of the biggest challenges" where disputes arise from conflicting claims and traditional evidence (screenshots, emails) can be edited or forged.

They position their solution as:

• Replacing manual, inconsistent human arbitration
• Using AI decisions grounded in cryptographically verified evidence
• Eliminating need for KYC
• Providing private dispute resolution
• Automating the entire escrow lifecycle from creation to settlement

The claim evolution shows a progression from identifying the problem (trust in online commerce) to proposing a technical solution (blockchain + AI + cryptography) to demonstrating implementation (full-stack system).

The description states TrustWise AI targets "online commerce" generally, including:

• Freelance work
• Peer-to-peer marketplaces
• Second-hand goods
• Digital services

It appears designed for:

• Buyers and sellers in online transactions who need dispute resolution
• Platforms that want to offer escrow services
• Users who want private, automated dispute resolution without KYC

The system is described as a "trust layer for online commerce" suggesting it's intended to be integrated into existing platforms rather than serving end-users directly.

No specific customer segments or personas are identified beyond general online commerce participants. The ICP appears to be platform providers or marketplace operators who want to offer escrow services with automated dispute resolution.

Not evidenced.

The description does not contain any information about:

• Revenue model
• Pricing structure
• Monetization approach
• Customer acquisition costs
• Unit economics
• Commercial relationships

All claims are self-reported and unverified. No evidence of actual business model or pricing exists in the provided description.

The description provides detailed technical implementation information:

The system is described as having "reliability features" including centralized error handling, request logging, retry mechanisms, and automated testing.

Not evidenced.

The description states this is a hackathon project submitted to the OpenAI 2026 hackathon. No evidence of:

• Revenue generation
• Customer adoption
• User base
• Product-market fit
• Market traction
• Commercial relationships
• Product usage metrics

The authors note they built "a full-stack application" and "successfully designed and implemented a complete end-to-end AI-powered escrow dispute resolution platform," but this is self-reported functionality without demonstration of actual use or impact.

Not evidenced.

The description does not contain:

• Information about existing competitors
• Market size or growth data
• Competitive positioning
• Differentiation from existing escrow services
• Market analysis
• Industry trends

No evidence of competitive landscape, market positioning, or how TrustWise AI compares to other dispute resolution platforms is provided.

1. What specific types of disputes have you tested the AI arbitration engine on? How does it handle edge cases?

1. Can you demonstrate how GPT-4o actually processes the verified evidence and generates structured decisions? What are the confidence scores like in practice?

1. How do you ensure that the AI doesn't hallucinate or produce inconsistent results when reasoning over shipment data and textual claims?

1. What happens if one party submits false information to the system despite cryptographic verification of shipment data?

1. Have you tested the system's resilience under failure conditions (verifier failure, AI failure, blockchain failure)?

1. How do you plan to transition from the Sepolia testnet to mainnet while maintaining security and functionality?

1. What is your roadmap for expanding beyond just shipment tracking evidence to other types of verifiable data?

1. How do you intend to scale this system beyond a hackathon prototype?

1. What are the specific use cases where this system would be superior to existing escrow services?

1. How do you plan to monetize this platform, and what is your go-to-market strategy?