SP
BravenNow
EVMbench: Evaluating AI Agents on Smart Contract Security
| USA | technology | ✓ Verified - arxiv.org

EVMbench: Evaluating AI Agents on Smart Contract Security

#EVMbench #AI agents #smart contract security #Ethereum #benchmark #vulnerability detection #blockchain auditing

📌 Key Takeaways

  • EVMbench is a new benchmark for evaluating AI agents on smart contract security tasks.
  • It assesses AI performance in identifying and mitigating vulnerabilities in Ethereum smart contracts.
  • The benchmark aims to standardize testing of AI-driven security tools in the blockchain domain.
  • Results could guide development of more reliable AI agents for automated smart contract auditing.

📖 Full Retelling

arXiv:2603.04915v1 Announce Type: cross Abstract: Smart contracts on public blockchains now manage large amounts of value, and vulnerabilities in these systems can lead to substantial losses. As AI agents become more capable at reading, writing, and running code, it is natural to ask how well they can already navigate this landscape, both in ways that improve security and in ways that might increase risk. We introduce EVMbench, an evaluation that measures the ability of agents to detect, patch,

🏷️ Themes

AI Evaluation, Blockchain Security

📚 Related People & Topics

Ethereum

Ethereum

Open-source blockchain computing platform

Ethereum is a decentralized blockchain with smart contract functionality. Ether (abbreviation: ETH) is the native cryptocurrency of the platform. Among cryptocurrencies, ether is second only to bitcoin in market capitalization.

View Profile → Wikipedia ↗

AI agent

Systems that perform tasks without human intervention

In the context of generative artificial intelligence, AI agents (also referred to as compound AI systems or agentic AI) are a class of intelligent agents distinguished by their ability to operate autonomously in complex environments. Agentic AI tools prioritize decision-making over content creation ...

View Profile → Wikipedia ↗

Entity Intersection Graph

Connections for Ethereum:

🌐 Bitcoin 3 shared
🌐 Bybit 1 shared
🌐 Dogecoin 1 shared
👤 Shiba Inu 1 shared
👤 Cryptocurrency 1 shared
View full profile

Mentioned Entities

Ethereum

Ethereum

Open-source blockchain computing platform

AI agent

Systems that perform tasks without human intervention
}
Original Source
--> Computer Science > Machine Learning arXiv:2603.04915 [Submitted on 5 Mar 2026] Title: EVMbench: Evaluating AI Agents on Smart Contract Security Authors: Justin Wang , Andreas Bigger , Xiaohai Xu , Justin W. Lin , Andy Applebaum , Tejal Patwardhan , Alpin Yukseloglu , Olivia Watkins View a PDF of the paper titled EVMbench: Evaluating AI Agents on Smart Contract Security, by Justin Wang and 7 other authors View PDF HTML Abstract: Smart contracts on public blockchains now manage large amounts of value, and vulnerabilities in these systems can lead to substantial losses. As AI agents become more capable at reading, writing, and running code, it is natural to ask how well they can already navigate this landscape, both in ways that improve security and in ways that might increase risk. We introduce EVMbench, an evaluation that measures the ability of agents to detect, patch, and exploit smart contract vulnerabilities. EVMbench draws on 117 curated vulnerabilities from 40 repositories and, in the most realistic setting, uses programmatic grading based on tests and blockchain state under a local Ethereum execution environment. We evaluate a range of frontier agents and find that they are capable of discovering and exploiting vulnerabilities end-to-end against live blockchain instances. We release code, tasks, and tooling to support continued measurement of these capabilities and future work on security. Subjects: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) Cite as: arXiv:2603.04915 [cs.LG] (or arXiv:2603.04915v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2603.04915 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Justin Wang [ view email ] [v1] Thu, 5 Mar 2026 07:59:14 UTC (1,156 KB) Full-text links: Access Paper: View a PDF of the paper titled EVMbench: Evaluating AI Agents on Smart Contract Security, by Justin Wang and 7 other authors View PDF HTML TeX Sour...
Read full article at source

Source

arxiv.org

More from USA

News from Other Countries

🇬🇧 United Kingdom

🇺🇦 Ukraine