--- title: GPT‑6 Cyber: Advancing the frontier of aligned... description: OpenAI released GPT‑6 Cyber, a cybersecurity model for finding and validating exploitable vulnerabilities while staying within authorized scope. It scores... canonical: https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents | daily.dev og:description: OpenAI released GPT‑6 Cyber, a cybersecurity model for finding and validating exploitable vulnerabilities while staying within authorized scope. It scores... og:url: https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj og:image: https://api.daily.dev/og/posts/BF8BMMAXJ.png og:image:alt: GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents **[OpenAI](https://daily.dev/sources/openai)** · 10 min read · 0 upvotes · 0 comments ## Summary OpenAI released GPT‑6 Cyber, a cybersecurity model for finding and validating exploitable vulnerabilities while staying within authorized scope. It scores 48.5% on ExploitGym, up from 32.1% for GPT‑5.6 Cyber, a 51% relative gain. It avoided exploiting out-of-scope services in 99.93% of tests without a deployment prompt, and 100% with one. The model found 81 novel high-severity Firefox vulnerabilities and developed working exploits for 948 of 1,724 reproduced Linux kernel vulnerabilities. OpenAI classifies it as Critical in cybersecurity under its Preparedness Framework, up from High. It's initially available via API and Codex to select Daybreak Red customers. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How does GPT-6 Cyber's exploit success rate compare to GPT-5.6 Cyber on ExploitGym? GPT-6 Cyber scores 48.5% on ExploitGym, up from 32.1% for GPT-5.6 Cyber, a roughly 16 percentage point gain and a 51% relative improvement. ExploitGym tests whether a model can exploit a specified vulnerability in a remote target. GPT-6 Cyber also achieved 40.3% recall on a new offensive web penetration testing benchmark, compared with 32.0% for GPT-6 Astra. _Track frontier model benchmark jumps like this one on daily.dev as AI reshapes offensive security work._ ### How many Firefox vulnerabilities did OpenAI's GPT-6 Cyber model find? GPT-6 Cyber found 81 novel Firefox vulnerabilities classified by Mozilla as high severity, and showed that 30 additional known issues were more severe than previously reported. OpenAI disclosed the findings along with proposed patches, and Mozilla began landing fixes shortly after receiving the reports, released in advisory mfsa2026-118. _Security teams patching Firefox-adjacent code can follow these disclosures on daily.dev._ ### What is OpenAI's Preparedness Framework risk classification for GPT-6 Cyber? OpenAI classifies GPT-6 Cyber as Critical in cybersecurity under its Preparedness Framework, an increase from the High classification given to GPT-5.6 Cyber. This reflects the model's stronger exploit development capabilities, including outperforming GPT-6 Astra on most cyber tasks and reproducing 1,724 Linux kernel vulnerabilities with 948 working exploits. _Developers weighing the risk of frontier AI cyber tools can follow classification changes like this on daily.dev._ --- Tags: [#cyber](https://daily.dev/tags/cyber), [#ai-agents](https://daily.dev/tags/ai-agents), [#openai](https://daily.dev/tags/openai), [#vulnerability](https://daily.dev/tags/vulnerability), [#ai-governance](https://daily.dev/tags/ai-governance) [View this post on daily.dev](https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents","url":"https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj"},"datePublished":"2026-11-05T18:51:40.053Z","dateModified":"2026-11-06T00:13:49.441Z","description":"OpenAI released GPT‑6 Cyber, a cybersecurity model for finding and validating exploitable vulnerabilities while staying within authorized scope. It scores...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c57ca44cfbc770ec9943b8c3bacd4572?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c57ca44cfbc770ec9943b8c3bacd4572?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"OpenAI","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"OpenAI","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/984b3632316f4afaa3d5213d5c64fe65","url":"https://daily.dev/sources/openai"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"cyber,ai-agents,openai,vulnerability,ai-governance","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"OpenAI","item":"https://daily.dev/sources/openai"},{"@type":"ListItem","position":3,"name":"GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/gpt-6-cyber-advancing-the-frontier-of-aligned-cybersecurity-agents-bf8bmmaxj#faq","mainEntity":[{"@type":"Question","name":"How does GPT-6 Cyber's exploit success rate compare to GPT-5.6 Cyber on ExploitGym?","acceptedAnswer":{"@type":"Answer","text":"GPT-6 Cyber scores 48.5% on ExploitGym, up from 32.1% for GPT-5.6 Cyber, a roughly 16 percentage point gain and a 51% relative improvement. ExploitGym tests whether a model can exploit a specified vulnerability in a remote target. GPT-6 Cyber also achieved 40.3% recall on a new offensive web penetration testing benchmark, compared with 32.0% for GPT-6 Astra. Track frontier model benchmark jumps like this one on daily.dev as AI reshapes offensive security work."}},{"@type":"Question","name":"How many Firefox vulnerabilities did OpenAI's GPT-6 Cyber model find?","acceptedAnswer":{"@type":"Answer","text":"GPT-6 Cyber found 81 novel Firefox vulnerabilities classified by Mozilla as high severity, and showed that 30 additional known issues were more severe than previously reported. OpenAI disclosed the findings along with proposed patches, and Mozilla began landing fixes shortly after receiving the reports, released in advisory mfsa2026-118. Security teams patching Firefox-adjacent code can follow these disclosures on daily.dev."}},{"@type":"Question","name":"What is OpenAI's Preparedness Framework risk classification for GPT-6 Cyber?","acceptedAnswer":{"@type":"Answer","text":"OpenAI classifies GPT-6 Cyber as Critical in cybersecurity under its Preparedness Framework, an increase from the High classification given to GPT-5.6 Cyber. This reflects the model's stronger exploit development capabilities, including outperforming GPT-6 Astra on most cyber tasks and reproducing 1,724 Linux kernel vulnerabilities with 948 working exploits. Developers weighing the risk of frontier AI cyber tools can follow classification changes like this on daily.dev."}}]} ```