OpenAI is taking a deeper dive into cryptocurrency safety with the launch of EVMbench, a brand new testing framework designed to measure how effectively synthetic intelligence can perceive and probably safe good contracts on Ethereum and comparable blockchains.
Good contracts, self-executing code deployed on blockchains similar to Ethereum, energy decentralized exchanges, lending protocols, and a variety of on-chain monetary purposes. These contracts usually can’t be modified as soon as deployed, making their vulnerabilities probably critical.
EVMbench is OpenAI’s try and see if trendy AI methods can play a task in stopping these issues. The benchmark was in-built collaboration with cryptocurrency funding agency Paradigm and leverages real-world good contract vulnerabilities beforehand found by means of audits and safety competitions.
The system measures efficiency throughout three key capabilities: figuring out safety bugs, exploiting these bugs in a managed setting, and fixing susceptible code with out violating contracts.
OpenAI says its objective is to ascertain clear requirements for evaluating AI methods in blockchain safety, particularly as decentralized finance continues to safe billions of {dollars} in person funds. The stakes for good contracts are solely going to extend.
“Good contracts usually safe greater than $100 billion in open supply crypto property. As AI brokers enhance their potential to learn, write, and execute code, it turns into more and more essential to measure that potential in economically significant environments and encourage the usage of AI methods defensively to audit and implement deployed contracts,” OpenAI stated in a weblog publish.
Learn extra: Most Influential Individual: Sam Altman
