Instaclaim

Instaclaim is a revolutionary platform that simplifies and automates the insurance claims process by integrating cutting-edge blockchain and AI technologies. By leveraging two AI agents—one on the client side and another on the insurance company’s server—the platform ensures a seamless, efficient, and transparent claims experience.

---

System Design

The platform integrates blockchain, AI, and secure data handling mechanisms to create a trustworthy and automated insurance claims process. Its architecture revolves around the following components:

• Client-Side AI Agent: Processes user input, encrypts sensitive data, and securely interacts with the server.
• Server-Side AI Agent: Verifies claims, processes policies, and automates payouts using smart contract logic.
• Lit Protocol: Ensures privacy and security during data transfer and execution of AI actions.
• Ethereum Attestation Service (EAS): Verifies authenticity of health records and claim approvals.

---

Video demo

https://www.youtube.com/watch?v=yfKAiwU7HaQ

Key Features

1. AI-Powered Automation

• Leverages Coinbase AgentKit to autonomously verify insurance policies and process claims.
• Handles claim approvals and denials efficiently with clear reasoning.
• Automates payouts directly to the user’s wallet upon claim approval.

2. Privacy-Preserving Data Handling

• Utilizes Lit Protocol to encrypt sensitive user data on the client side.
• Decrypts and processes data securely on the server using Lit Actions.

3. Trust Through Attestations

• Integrates Ethereum Attestation Services (EAS) to authenticate health reports submitted by users.
• Ensures that only hospital-verified documents are considered valid for claims.

4. Blockchain-Backed Security

• Combines blockchain’s immutability with AI’s efficiency to deliver a tamper-proof solution.
• Enhances transparency and trust between clients and insurance providers.

---

Ethereum Attestation Service (EAS)

EAS is used to attest SHA256 hashes of health records along with users’ wallet addresses. This ensures access control and authenticity for sensitive data.

Workflow:

1. Hospital Attestation: Trusted partner hospitals create attestations containing health record hashes and user wallet addresses.
2. Claim Verification:
   • When a user submits a claim, their data and wallet address are matched against these attestations.
   • If the hash matches, access is granted to process the claim.

Schema and Transaction Details:

• Schema ID for Hospital Records: View on Sepolia

• Example Attestation ID: 0x2238f03eb415824d725e809253f0c3928e87898cca44eb4121e4ff2815ceccb1

• Schema ID for Insurance Records: View on Base Sepolia

• Example Attestation ID: 0x2eb5ba3523b17555bd8e6aba7f593e71b57520f329b9fa55286e0d11c932f352

---

Deployment of Coinbase Agent

The Coinbase Agent, built using AgentKit, is trained with the insurance company’s policies and operates autonomously to:

1. Approve Claims: Automatically send payouts to the user’s wallet.
2. Deny Claims: Provide clear reasons for denial.

Example API:

The agent is hosted on Replit using a Python Flask container with an open API for interactions.

Replit Repo: View Repo

API Usage:

curl -X POST https://onchain-agent-demo-backend-architdabral123.replit.app/api/chat \
-H "Content-Type: application/json" \
-d '{"input": "Patricia experienced a severe allergic reaction and went to an emergency room for treatment. The treatment was deemed medically necessary by the attending physician. No coverage limits have been exceeded. Cost of Treatment: 0.002 Wallet: 0xfcd1e86925C9c066d31AacC78c9e7De32b4574Ae", "conversation_id": 0}'

Use of Lit Protocol

The Lit Protocol is the backbone of the platform’s security, enabling privacy-preserving interactions and ensuring that sensitive data is handled securely. It is utilized both on the client and server sides to manage encryption, decryption, and access control.

Workflow of Lit Protocol Integration

1. Client-Side Operations

• AI Agent Execution: The first AI agent is invoked as a Lit Action to process the user’s input.
• Data Encryption: The response from the AI agent is encrypted using another Lit Action, ensuring that sensitive data remains private during transmission.
• Secure Transfer: The encrypted response is sent to the server, ensuring that only authorized entities can decrypt and process the data.

2. Server-Side Operations

• Data Decryption: Upon receiving the encrypted response, the server decrypts it using Access Control Conditions (ACC). These conditions ensure that only the server’s authorized wallet has access to the data.
• Claim Processing: The server-side AI agent then processes the decrypted data to verify the claim and execute the appropriate action (approve or deny).

Key Advantages of Using Lit Protocol

1. Privacy Protection: All sensitive user data is encrypted before transmission and is accessible only to authorized parties.
2. Trustless Security: Encryption and decryption are handled through Lit Actions, providing a trustless execution environment.
3. Granular Access Control: Access Control Conditions (ACC) enable precise and conditional access to data based on predefined rules.

Lit Actions in the Project

1. Client-Side Lit Actions:

   • AI Agent Invocation: Executes the client-side AI agent as a Lit Action.
   • Response Encryption: Encrypts the agent's response for secure transmission.

2. Server-Side Lit Actions:

   • Data Decryption: Decrypts client-provided data based on ACC.
   • Secure Processing: Ensures all claim-related operations are conducted in a secure and trusted environment.

---

By leveraging the Lit Protocol, the platform achieves a robust, privacy-preserving solution that guarantees secure data handling and efficient claim processing.