Post #11682

OpenAI partners with Paradigm to launch AI agent tool for smart contract security To test AI agents' capabilities, OpenAI and Paradigm have introduced EVMbench, a benchmark evaluating the ability of AI agents to detect, patch, and exploit high-severity smart contract vulnerabilities. The unveiling of EVMbench comes on the heels of DeFi lending protocol Moonwell suffering an exploit this month. The episode involved vulnerable code written with AI assistance. EVMbench will offer different "capability modes," including detecting vulnerabilities, modifying contracts, and eliminating exploitability, plus testing an AI agent's ability to "execute end-to-end fund-draining attacks against deployed contracts on a sandboxed blockchain environment. EVMbench reportedly draws on 120 curated vulnerabilities from 40 audits, including sponsored open-code audit competitions and the security auditing process for the Tempo, the Layer 1 blockchain co-developed by Paradigm and Stripe. 🔗openai.com/index/introducing-evmbench @TOP7ICO