Ethereum’s co-founder, Vitalik Buterin, believes that artificial intelligence (AI) could play a vital role in addressing one of Ethereum’s major technical risks: hidden bugs within its code.
In a post on X on February 18th, Buterin expressed his enthusiasm for AI-powered auditing, which can help identify and fix buggy code in the Ethereum network. He considers this to be the most significant technical risk to the network.
Buterin’s comments come as Ethereum approaches the implementation of its highly anticipated Dencun upgrade, scheduled for launch on March 13th. Although Dencun was successfully implemented on the Goerli testnet on January 17th, a bug in Prsym caused a four-hour delay in finalizing the network on the testnet. Upgrades to the Ethereum network are crucial for its long-term roadmap.
However, not everyone shares Buterin’s confidence in using AI to detect bugs in Ethereum-based code. In July 2023, OpenZeppelin conducted a series of experiments using OpenAI’s GPT-4 to identify security issues in Solidity smart contracts, the native language of Ethereum code. GPT-4 successfully identified vulnerabilities in 20 out of 28 challenges during these experiments. However, there were instances where GPT-4 either failed to detect flaws or even invented vulnerabilities that didn’t exist.
Similarly, Kang Li, the chief security officer at CertiK, told Cointelegraph that using AI-powered tools in coding often leads to more security issues than it solves. Li recommends using AI assistants only as aids to experienced coders, as they can quickly explain the meaning of a line of code.
While Buterin remains optimistic about the future of AI, he has previously cautioned developers to be careful when implementing AI with blockchain technology, especially when using it alongside high-risk applications like oracles. He warns that if an AI oracle used in a prediction market or a stablecoin is vulnerable to attacks, a substantial amount of money could be lost instantly.
In conclusion, while AI shows promise in detecting bugs in Ethereum’s code, there are concerns about its reliability. It is recommended that AI tools be used as assistants to experienced coders, and caution should be exercised when deploying AI in high-risk applications within the blockchain.