{"version":1,"type":"story","url":"https://digestai.news/story/openai-releases-gpt-6-cyber-claims-81-novel-firefox-vulnerabilities","json":"https://digestai.news/story/openai-releases-gpt-6-cyber-claims-81-novel-firefox-vulnerabilities.json","markdown":"https://digestai.news/story/openai-releases-gpt-6-cyber-claims-81-novel-firefox-vulnerabilities.md","slug":"openai-releases-gpt-6-cyber-claims-81-novel-firefox-vulnerabilities","headline":"OpenAI launches GPT-6 Cyber, finds 81 Firefox flaws and 948 Linux exploits","summary":"OpenAI has released GPT-6 Cyber, a model trained to find and validate exploitable vulnerabilities while staying within researcher-defined scope. The company reports the model scores 48.5% on the ExploitGym benchmark, up from 32.1% for GPT-5.6 Cyber, and avoided out-of-scope actions in 99.93% of tests without safety prompts and 100% with them. OpenAI classifies the model as Critical under its Preparedness Framework.\n\nIn real-world testing, GPT-6 Cyber identified 81 previously unknown high-severity vulnerabilities in Firefox, with Mozilla releasing fixes for many. It also reproduced 1,724 Linux kernel vulnerabilities and developed working exploits for 948 of them, reporting 304 critical findings to maintainers. The Linux campaign cost $1,213,430 in compute. Security teams at Cisco, Calif, CodeRabbit, and Palo Alto Networks are using the model, which is initially available to select Daybreak Red customers via API and Codex.\n\nThe Decoder notes all benchmark figures come from OpenAI with no independent testing published. It also reports OpenAI's claim that agents used at least ten novel attack techniques unknown to external kernel experts. Note.com adds the release coincided with a 12% drop in the SOX semiconductor index and METR's finding that Anthropic's development speed increased 1.9x.","keyPoints":["GPT-6 Cyber scores 48.5% on ExploitGym vs 32.1% for GPT-5.6 Cyber; classified Critical in Preparedness Framework","Model found 81 novel Firefox vulnerabilities and developed exploits for 948 Linux kernel flaws","Available to select Daybreak Red customers; Linux research campaign cost $1,213,430 in compute"],"whyItMatters":"Demonstrates AI can autonomously find and exploit vulnerabilities at scale while respecting scope boundaries, shifting cybersecurity toward automated defense but raising dual-use concerns for critical infrastructure.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["OpenAI","Mozilla","Linux","Cisco","Calif","CodeRabbit"],"models":["GPT-6 Cyber","GPT-5.6 Cyber","GPT-6 Astra","GPT-5.6 Sol","GLM"],"people":[]},"firstPublishedAt":"2026-11-05T17:15:02Z","updatedAt":"2026-11-05T23:00:00Z","sourceCount":15,"hasPrimarySource":true,"sources":[{"outlet":"openai.com","title":"OpenAI has just announced GPT‐6 Cyber, a major step up from GPT‐5.6 Cyber and outperforms GPT‐6 Astra on most cyber tasks","url":"https://openai.com/index/introducing-gpt-6-cyber","publishedAt":"2026-11-05T17:13:30Z","type":"primary","primary":true,"lead":true},{"outlet":"note.com","title":"AI Evolution at 'Double Speed,' Stock Prices Plunge 5.5%: The Gap Between Technology and Society Shown by OpenAI's New Cyber AI | November 6, 2026","url":"https://note.com/ripe_crab6964/n/n8b3087a723a2?hl=en","publishedAt":"2026-11-05T23:00:00Z","type":"press","primary":false,"lead":false},{"outlet":"businesstoday.in","title":"An AI model that can hack? OpenAI's new GPT-6 Cyber can find and exploit security flaws","url":"https://businesstoday.in/technology/artificial-intelligence/story/an-ai-model-that-can-hack-openais-new-gpt-6-cyber-can-find-and-exploit-security-flaws-566106-2026-11-06","publishedAt":"2026-11-05T21:34:00Z","type":"press","primary":false,"lead":false},{"outlet":"note.com","title":"Global AI Brief | OpenAI Releases 'Critical' Capability GPT-6 Cyber; Panasonic Achieves 65% Success Rate with Tactile Robot; PFN Releases Translation LLM","url":"https://note.com/china_case/n/n28b23e15f57c?hl=en","publishedAt":"2026-11-05T21:29:00Z","type":"press","primary":false,"lead":false},{"outlet":"timesofindia.indiatimes.com","title":"OpenAI launches GPT-6 Cyber, finds 81 high-severity Firefox flaws and 948 Linux exploits","url":"https://timesofindia.indiatimes.com/technology/tech-news/openai-launches-gpt-6-cyber-finds-81-high-severity-firefox-flaws-and-948-linux-exploits/articleshow/135595138.cms","publishedAt":"2026-11-05T21:26:00Z","type":"press","primary":false,"lead":false},{"outlet":"The Decoder","title":"OpenAIs GPT-6 Cyber can both break into systems and propose how to fix them","url":"https://the-decoder.com/openais-gpt-6-cyber-can-both-break-into-systems-and-propose-how-to-fix-them","publishedAt":"2026-11-05T20:36:25Z","type":"press","primary":false,"lead":false},{"outlet":"note.com","title":"AI News Morning Edition | November 6, 2026 (Fri) 6:00 AM Edition: OpenAI releases 'Critical' level cyber AI, Anthropic development pace 1.9x faster!","url":"https://note.com/gifted_ixora801/n/n61e18c7fcc37?hl=en","publishedAt":"2026-11-05T19:09:00Z","type":"press","primary":false,"lead":false},{"outlet":"TechCrunch AI","title":"OpenAI released a new cybersecurity model. Is that enough?","url":"https://techcrunch.com/2026/11/05/openai-released-a-new-cybersecurity-model-is-that-enough","publishedAt":"2026-11-05T19:06:12Z","type":"press","primary":false,"lead":false},{"outlet":"ciso.economictimes.indiatimes.com","title":"OpenAI launches GPT-6 Cyber with 'Critical' cybersecurity rating","url":"https://ciso.economictimes.indiatimes.com/news/vulnerabilities-exploits/openai-launches-gpt-6-cyber-with-critical-cybersecurity-rating/135590741","publishedAt":"2026-11-05T18:32:00Z","type":"press","primary":false,"lead":false},{"outlet":"Unite.AI","title":"OpenAI Launches GPT-6 Cyber With Restricted Access and Scope Controls","url":"https://unite.ai/openai-launches-gpt-6-cyber-with-restricted-access-and-scope-controls","publishedAt":"2026-11-05T17:38:03Z","type":"press","primary":false,"lead":false},{"outlet":"cyberscoop.com","title":"OpenAI says GPT-6 Cyber will first launch to a limited number of customers in its Daybreak Red program and has found 81+ \"novel\" vulnerabilities in Firefox","url":"https://cyberscoop.com/openai-gpt-6-cyber-security-model","publishedAt":"2026-11-05T17:15:02Z","type":"press","primary":false,"lead":false},{"outlet":"theverge.com","title":"OpenAI launches GPT-6 Cyber, which it says outperforms GPT-6 Astra on cyber benchmarks, is better at finding vulnerabilities, and is less likely to go rogue","url":"https://theverge.com/ai-artificial-intelligence/1027599/openai-gpt-6-cyber-launch","publishedAt":"2026-11-05T17:06:54Z","type":"press","primary":false,"lead":false},{"outlet":"digit.in","title":"OpenAI launches GPT-6 Cyber AI model, says it found 81 Firefox vulnerabilities","url":"https://digit.in/news/general/openai-launches-gpt-6-cyber-ai-model-says-it-found-81-firefox-vulnerabilities.html","publishedAt":"2026-11-05T16:00:00Z","type":"press","primary":false,"lead":false},{"outlet":"siliconangle.com","title":"OpenAI launches GPT-6 Cyber, an Astra model built to keep vulnerability research in scope","url":"https://siliconangle.com/2026/11/05/openai-launches-gpt-6-cyber-a-gpt-6-astra-model-built-to-keep-vulnerability-research-in-scope","publishedAt":"2026-11-05T14:51:00Z","type":"press","primary":false,"lead":false},{"outlet":"cryptobriefing.com","title":"OpenAI brings GPT-5.6 Cyber to select Daybreak Red customers","url":"https://cryptobriefing.com/openai-gpt-6-cyber-daybreak-red","publishedAt":"2026-11-05T09:16:00Z","type":"press","primary":false,"lead":false}],"sourceNotes":{"agree":"All four sources report OpenAI launched GPT-6 Cyber, a cybersecurity model that found 81 Firefox vulnerabilities and developed exploits for 948 Linux kernel flaws, scoring 48.5% on ExploitGym.","differ":["Note.com links the release to a 12% SOX index drop and METR's 1.9x Anthropic speed finding, which other sources omit","Digit.in lists Palo Alto Networks as a user; OpenAI.com and The Decoder name Cisco, Calif, and CodeRabbit"]},"discussions":[{"site":"Reddit","url":"https://www.reddit.com/r/singularity/comments/1xoo9ob/openai_has_just_announced_gpt6_cyber_a_major_step/","points":null}],"thread":{"title":"OpenAI Powers Corporate Cybersecurity Automation","url":"https://digestai.news/thread/block-uses-openai-daybreak-to-boost-security-triage-throughput-by-50","storyCount":4},"cite":{"text":"Digest AI, \"OpenAI launches GPT-6 Cyber, finds 81 Firefox flaws and 948 Linux exploits\", 5 November 2026, https://digestai.news/story/openai-releases-gpt-6-cyber-claims-81-novel-firefox-vulnerabilities","publisher":"Digest AI","title":"OpenAI launches GPT-6 Cyber, finds 81 Firefox flaws and 948 Linux exploits","datePublished":"2026-11-05T17:15:02Z","url":"https://digestai.news/story/openai-releases-gpt-6-cyber-claims-81-novel-firefox-vulnerabilities"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}