OpenAI is diving deeper into cryptographic security with the launch of EVMbench, a new testing framework designed to measure how well artificial intelligence can understand and potentially secure smart contracts on Ethereum and similar blockchains.
Smart contracts, the self-executing code deployed on blockchains like Ethereum, underpin decentralized exchanges, lending protocols, and a wide range of on-chain financial applications. Since these contracts are generally immutable once deployed, the vulnerabilities can be serious.
EVMbench is OpenAI’s attempt to see if modern AI systems are up to the task of helping prevent these problems. Built in collaboration with crypto investment firm Paradigm, the benchmark draws on real-world smart contract vulnerabilities previously discovered during security audits and competitions.
The system measures performance based on three primary capabilities: identifying security bugs, exploiting those bugs in a controlled environment, and fixing vulnerable code without breaking contracts.
OpenAI says the goal is to establish a clear standard for evaluating AI systems in blockchain security, especially as decentralized finance continues to secure billions of dollars of user funds. The stakes for smart contracts are only increasing.
“Smart contracts regularly secure over $100 billion in open source crypto assets. As AI agents improve their ability to read, write, and execute code, it becomes increasingly important to measure their capabilities in economically meaningful environments and encourage defensive use of AI systems to audit and enforce deployed contracts,” OpenAI wrote in a blog post.
Read more: Most influential: Sam Altman
