OpenAI and Paradigm Launch EVMbench to Test AI Agents on Smart Contract Security
A new benchmark from OpenAI and Paradigm evaluates how well AI agents can detect, exploit, and patch vulnerabilities in Ethereum smart contracts — bringing rigorous evaluation to the fast-growing AI-for-crypto-security space.
OpenAI has partnered with crypto research firm Paradigm to release EVMbench, a benchmark designed to evaluate AI agents' ability to find, exploit, and remediate vulnerabilities in Ethereum Virtual Machine smart contracts. As @tradeguru reported, the benchmark tests leading models including GPT, Claude, and Gemini across a range of real-world vulnerability classes.
Unlock the full briefing
Get every story in today's briefing, the full archive, and the daily AI intelligence brief.
All stories today
Full archive
Daily brief
Cancel anytime. Payments powered by Stripe.