Etherscan, the Ethereum block explorer and analytics platform, has unveiled an innovative tool called "Code Reader" on June 19. Powered by artificial intelligence (AI), Code Reader allows users to retrieve and interpret the source code of specific contract addresses. By leverage ing OpenAI's large-scale language model, the tool generates responses that provide insights into the contract's source code file. Users can gain a deeper understanding of the code, obtain a comprehensive list of smart contract functions, and explore how underlying contracts interact with decentralized applications. Notably, Code Reader requires a valid OpenAI API key and ensures the privacy of users' keys by not storing them.
While the field of AI continues to thrive, concerns have been raised about the practicality of current AI models. Foresight Ventures, a venture capital firm based in Singapore, highlighted the forthcoming battle for computing power resources. As demand for training large AI models grow, Researchers acknowledge the limitations of current prototypes, including challenges in data synchronization, network optimization, and data privacy and security. For instance, training a model with 175 billion parameters would require significant storage space, and distributed train ing would necessitate frequent data transfers and updates between computing nodes. Despite the excitement around larger models, the report emphasizes that small AI models remain a more viable option in many cases and should not be disregarded prematurely.
Overall, Etherscan's Code Reader introduces an AI-powered solution to facilitate the interpretation of Ethereum contract source code. However, the wider AI landscape continues to face challenges in resource allocation and practical implementation, highlighting the importance of considering the practicality and scalability of AI models .
