Generative AI Blockchain Security Analysis System

Effective And Fast.

Comment

Inspiration

The inspiration for our Generative AI Blockchain Security Analysis System stemmed from the alarming rise in blockchain fraud. The Crypto Crime Report 2023 from Chainalysis reported a staggering $20.6 billion lost in 2022 alone. Traditional prevention methods are slow and cumbersome, often requiring constant vigilance on social media for updates. We envisioned a more efficient, user-friendly solution leveraging the power of generative AI to provide real-time, automated security analysis.

What it does

Our system offers two main features: Smart Contract Vulnerability Analysis: Users can input an ETH smart contract address or the Solidity version and contract code directly. The system retrieves the contract details, analyzes potential vulnerabilities using OpenAI’s LLM, and generates a comprehensive report. Fraud Transaction Analysis: Users input a transaction hash from Ethereum, Polygon, or Avalanche chains. The system gathers transaction data through Chainlink functions, analyzes it with OpenAI’s LLM, and provides a detailed fraud assessment report.

How we built it

Research: Generative AI: We explored capabilities of Gemini and ChatGPT to understand how they could be leveraged for security analysis. Blockchain Security: In-depth research on blockchain security threats and prevention techniques. Blockchain Applications: Knowledge of smart contracts, oracles, Chainlink functions, and QuickNode IPFS was crucial. On-Chain Data: Focused on data from Polygon, Avalanche, and Ethereum chains. Language and Framework: Backend: Python and Node.js Frontend: Angular Team: Kenneth Ng Hugo Leung Horace

Challenges we ran into

  1. Fetching On-Chain Data: Integrating on-chain data with OpenAI API posed technical challenges.
  2. On-Chain and Off-Chain Data Operations: Ensuring seamless operation between blockchain data and AI analysis.
  3. QuickNode IPFS Application: Implementing and managing the QuickNode IPFS system for storing analysis reports.
  4. Generative AI Research: Identifying the best approaches for utilizing generative AI in security analysis.

Accomplishments that we're proud of

  1. Successfully integrating OpenAI’s LLM for real-time smart contract and transaction analysis.
  2. Developing a user-friendly web application that simplifies blockchain security.
  3. Efficiently handling large volumes of on-chain and off-chain data.
  4. Implementing a robust storage solution using QuickNode IPFS.

What we learned

  1. The intricacies of blockchain security and the importance of real-time analysis.
  2. Effective use of generative AI to enhance security measures.
  3. Technical challenges and solutions in integrating on-chain data with AI systems.
  4. Importance of user-friendly design in cybersecurity applications.

What's next for Generative AI Blockchain Security Analysis System

  1. Expansion to More Chains: Including additional blockchain networks to broaden our analysis capabilities.
  2. Cross Chain Analysis: Develop analysis for cross chain analysis with different chains
  3. Enhanced AI Models: Continuously improving AI models for more accurate and comprehensive security analysis.
  4. User Interface Improvements: Further refining the user interface to enhance usability and accessibility.
  5. Partnerships: Collaborating with blockchain security firms and organizations to expand the reach and impact of our system.

Built With

Share this project:

Updates