OpenAI, along with Paradigm and Ottersec, has released the EVMbench research paper, looking at how well different AI models can detect, patch up, and exploit smart contract vulnerabilities.

OpenAI and Paradigm launched EVMbench, a tool that tests how capable AI agents are at finding and fixing smart contract vulnerabilities.

OpenAI launches its OpenAI Smart Contract Benchmark called EVMbench with Paradigm to test AI agents in crypto audits.

EVMbench is OpenAI’s attempt to see whether modern AI systems are up to the task of helping prevent smart contract issues.

Moonwell’s $1.78 million oracle mispricing exploit is reigniting debate over “vibe-coded” smart contracts and how AI tools like Claude Opus 4.6 should be governed in DeFi development.

DeFi lending protocol Moonwell lost $1.78 million after AI code from Claude Opus 4.6 led to an oracle configuration error.

While AI-powered tools are getting better at coding, errors still occur and can cost millions, as an example from this week has already demonstrated.

Claude Opus 4.6 vulnerable code caused a $1.78M Moonwell exploit after an oracle error mispriced cbETH, raising fresh concerns over AI-written Solidity.

Claude Opus 4.6 linked to a $1.78M DeFi hack after cbETH mispricing. Auditor Pashov and SlowMist cite oracle formula vulnerability concerns.

NearSpace Education (NSE) has been awarded a NASA TEAM II STEM Innovator grant to erase the line between classroom ...