{"id":10774,"date":"2026-02-19T10:03:34","date_gmt":"2026-02-19T10:03:34","guid":{"rendered":"https:\/\/serisec.com\/index.php\/2026\/02\/19\/openai-launches-evmbench-to-detect-patch-and-exploit-vulnerabilities-in-blockchain-environments\/"},"modified":"2026-02-19T10:03:34","modified_gmt":"2026-02-19T10:03:34","slug":"openai-launches-evmbench-to-detect-patch-and-exploit-vulnerabilities-in-blockchain-environments","status":"publish","type":"post","link":"https:\/\/serisec.com\/index.php\/2026\/02\/19\/openai-launches-evmbench-to-detect-patch-and-exploit-vulnerabilities-in-blockchain-environments\/","title":{"rendered":"OpenAI Launches EVMbench to Detect, Patch, and Exploit Vulnerabilities in Blockchain Environments"},"content":{"rendered":"

OpenAI Launches EVMbench to Detect, Patch, and Exploit Vulnerabilities in Blockchain Environments
\n \t

\n
<\/BR>
\n
\n \t

\n
<\/BR><\/p>\n

\n

OpenAI, in collaboration with crypto investment firm Paradigm, has introduced EVMbench, a new benchmark designed to evaluate the ability of AI agents to detect, patch, and exploit high-severity vulnerabilities in smart contracts<\/a>.<\/p>\n

The release marks a significant step in measuring AI capabilities within economically consequential environments, as smart contracts routinely secure over $100 billion in open-source crypto assets.<\/p>\n

EVMbench draws on 120 curated vulnerabilities sourced from 40 security audits, with the majority derived from open code audit competitions on platforms such as Code4rena.<\/p>\n

The benchmark also incorporates vulnerability scenarios from the security auditing process of the Tempo blockchain, a purpose-built Layer 1 designed for high-throughput stablecoin payments, extending EVMbench\u2019s scope into payment-oriented smart contract code an area where agentic stablecoin transactions are expected to grow substantially.<\/p>\n

Three Evaluation Modes<\/strong><\/h2>\n

EVMbench evaluates AI agents across three distinct capability modes, each targeting a different phase of the smart contract security lifecycle.<\/p>\n

\n\n\n\n\n\n\n\n
Mode<\/th>\nDescription<\/th>\n<\/tr>\n<\/thead>\n
Detect<\/td>\nAgents audit a smart contract repository and are scored on recall of ground-truth vulnerabilities and associated audit rewards<\/td>\n<\/tr>\n
Patch<\/td>\nAgents modify vulnerable contracts while preserving intended functionality, verified through automated tests and exploit checks<\/td>\n<\/tr>\n
Exploit<\/td>\nAgents execute end-to-end fund-draining attacks against deployed contracts in a sandboxed blockchain environment, graded via transaction replay and on-chain verification<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n

To support reproducible evaluation, OpenAI developed a Rust-based harness that deploys contracts deterministically and restricts unsafe RPC methods. All exploit tasks run in an isolated local Anvil environment rather than on live networks.<\/p>\n

Frontier model performance on EVMbench reveals clear behavioral differences across task types. In the exploit mode, GPT\u20115.3\u2011Codex achieved a score of 72.2%, a substantial improvement over GPT\u20115, which scored 31.9% approximately six months prior.<\/p>\n

Agents consistently perform best on exploit tasks, where the objective is explicit: drain funds and iterate until successful. Detect and patch modes remain harder, with agents sometimes stopping after identifying a single vulnerability rather than completing a full audit, and struggling to remove subtle flaws without breaking existing contract functionality.<\/p>\n

OpenAI acknowledged that EVMbench<\/a> does not fully reflect the difficulty of real-world smart contract security, and that its grading system cannot currently distinguish between true vulnerabilities and false positives when agents find issues beyond the human-auditor baseline.<\/p>\n

Alongside the benchmark release, OpenAI committed $10 million in API credits through its Cybersecurity Grant Program to accelerate defensive security research, particularly for open-source software and critical infrastructure.<\/p>\n

The company also announced the expansion of Aardvark<\/a>, its security research agent, through a private beta program. EVMbench\u2019s tasks, tooling, and evaluation framework have been released publicly to support continued research into AI-driven cyber capabilities.<\/p>\n

Follow us on Google News<\/a>, LinkedIn<\/a>, and X<\/a> for daily cybersecurity updates. Contact us<\/a> to feature your stories.<\/strong><\/p>\n

The post OpenAI Launches EVMbench to Detect, Patch, and Exploit Vulnerabilities in Blockchain Environments<\/a> appeared first on Cyber Security News<\/a>.<\/p>\n<\/div>\n

\t

\n
<\/BR>
\n Guru Baran
\n \t

\n
<\/BR>
\nGo to cyber-security-news<\/a>
\n \t

\n
<\/BR><\/p>\n","protected":false},"excerpt":{"rendered":"

OpenAI Launches EVMbench to Detect, Patch, and Exploit Vulnerabilities in Blockchain Environments OpenAI, in collaboration with crypto investment firm Paradigm, has introduced EVMbench, a new benchmark designed to evaluate the ability of AI agents to detect, patch, and exploit high-severity vulnerabilities in smart contracts. The release marks a significant step in measuring AI capabilities within […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[129,63],"tags":[130],"class_list":["post-10774","post","type-post","status-publish","format-standard","hentry","category-cyber-security","category-cyber-security-news","tag-cyber-security-news"],"_links":{"self":[{"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/posts\/10774"}],"collection":[{"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/comments?post=10774"}],"version-history":[{"count":0,"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/posts\/10774\/revisions"}],"wp:attachment":[{"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/media?parent=10774"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/categories?post=10774"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/serisec.com\/index.php\/wp-json\/wp\/v2\/tags?post=10774"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}