AINewsOpenAIOpenAI says its new cybersecurity model is better at not hacking other companiesClassified as \u2018Critical,\u2019 GPT-6 Cyber outperforms GPT-6 Astra on cyber-related benchmarks.Classified as \u2018Critical,\u2019 GPT-6 Cyber outperforms GPT-6 Astra on cyber-related benchmarks.by Emma RothNov 5, 2026, 5:00 PM UTCShareGiftImage: The VergeEmma Roth is a news writer who covers the streaming wars, consumer tech, crypto, social media, and much more. Previously, she was a writer and editor at MUO.OpenAI is launching GPT-6 Cyber, the latest version of its cybersecurity AI model, which it says is better at finding vulnerabilities and less likely to go rogue and hack other platforms in the process. OpenAI says that during testing, GPT-6 Cyber went out of bounds less than 1 percent of the time, down from around 5 percent with GPT-5.6 Cyber.The evaluation, called ExploitGym HoneyPot, is modeled after the Hugging Face Incident, in which OpenAI\u2019s AI agents breached the open-source AI community to find answers to a problem they were tasked with solving. \u201cWith the 6 Cyber model, we were able to make huge improvements on this evaluation,\u201d Fouad Matin, the head of OpenAI\u2019s cyber preparedness and safety team, said during a press briefing. \u201cIt\u2019s over 99 percent of the time doing what you ask.\u201dBut evaluation awareness, or when an AI system recognizes that it\u2019s being tested and changes its behavior in response, remains a concern. \u201cEvaluation awareness is super top of mind for us,\u201d Matin said. \u201cOften when we do see evaluation awareness, you\u2019ll see it in the model\u2019s chain of thought.\u201d Matin added that OpenAI is trying to prioritize production-like evaluations, where the model is interacting with real codebases with a real objective to lessen the risk of awareness.Following a wave of AI cybersecurity incidents, OpenAI has raced to strengthen its safeguards and lock down its most powerful models. The company briefly paused training of its frontier AI models, and it now restricts their access to the internet.GPT-6 Cyber is classified as Critical in cybersecurity on OpenAI\u2019s Preparedness Framework, which indicates that it \u201ccan identify and develop functional zero-day exploits of all severity levels in many hardened real-world critical systems without human intervention.\u201d GPT-5.6 Cyber was previously classified as High under the framework.GPT-6 Cyber outperforms GPT-6 Astra and GPT-5.6 Sol on ExploitGym.<\/p>\n
Image: OpenAIGPT-6 Cyber outperforms GPT-6 Astra, OpenAI\u2019s flagship Critical-level model, on most cyber-related tasks. In ExploitGym, an AI benchmark that measures whether an AI model can exploit vulnerabilities, GPT-6 Cyber scored 48.5 percent, about 16 points higher than GPT-5.6 Cyber\u2019s 32.1 percent and a few points higher than GPT-6 Astra\u2019s 42.4 percent score. The model helped find 81 previously unknown vulnerabilities in Firefox, and it played a role in hardening OpenAI\u2019s own infrastructure for training AI models, OpenAI says.Additionally, GPT-6 Cyber comes with stronger safeguards \u201cto detect and disrupt misuse\u201d while better adhering to user instructions. OpenAI says it\u2019s using new checks to block \u201chigh-risk activity,\u201d including attacks on critical infrastructure like power plants and water treatment facilities.RelatedInside the suddenly explosive world of AI safetyOpenAI now trains its frontier AI models with no internet accessGPT-6 Cyber is rolling out to select members of OpenAI\u2019s Daybreak Red program, which is available to companies and organizations using its models for \u201cauthorized defensive cybersecurity work.\u201d It\u2019s expanding GPT-6 Cyber to all Daybreak Red customers in the coming weeks.Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.Emma RothAINewsOpenAIMost PopularAnker\u2019s new window-mounted camera monitors the outside of your home from within itSam Altman addresses RAM comments: People should \u2018put as much as they want in their computers\u2019Those Jeff Bezos hair transplant photos are very fakeMy therapist thinks I should chuck the Amazon Bee Frontier out a windowDiscord is ditching Electron and its new app uses less RAMAdvertiser Content FromThis is the title for the native ad<\/p>\n","protected":false},"excerpt":{"rendered":"
AINewsOpenAIOpenAI says its new cybersecurity model is better at not hacking other companiesClassified as \u2018Critical,\u2019 GPT-6 Cyber outperforms GPT-6 Astra on cyber-related benchmarks.Classified as \u2018Critical,\u2019 GPT-6 Cyber outperforms GPT-6 Astra on cyber-related benchmarks.by Emma RothNov\u2026<\/p>\n","protected":false},"author":1,"featured_media":34227,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[],"class_list":["post-34226","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology"],"_links":{"self":[{"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/posts\/34226","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/comments?post=34226"}],"version-history":[{"count":0,"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/posts\/34226\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/media\/34227"}],"wp:attachment":[{"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/media?parent=34226"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/categories?post=34226"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/demo.tcjhomeaccents.com\/wp\/wp-json\/wp\/v2\/tags?post=34226"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}