{"id":1835,"date":"2025-12-11T10:42:29","date_gmt":"2025-12-11T10:42:29","guid":{"rendered":"https:\/\/www.tsfactory.com\/forums\/?p=1835"},"modified":"2025-12-11T10:44:14","modified_gmt":"2025-12-11T10:44:14","slug":"exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says","status":"publish","type":"post","link":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/","title":{"rendered":"Exclusive: New OpenAI models likely pose &#8220;high&#8221; cybersecurity risk, company says"},"content":{"rendered":"<div class=\"gtmView mb-5 sm:mb-10\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-event-name=\"story_view\" data-vars-deprecated-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-deprecated-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-deprecated-category=\"story\" data-vars-deprecated-sub-category=\"story\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-latitude=\"55.95\" data-vars-longitude=\"-3.2\" data-vars-postal-code=\"EH1\" data-vars-product=\"core\" data-vars-page-type=\"story\" data-vars-sub-product=\"national\" data-gtm-vis-first-on-screen48430672_37=\"30234\" data-gtm-vis-total-visible-time48430672_37=\"1000\" data-gtm-vis-first-on-screen48430672_468=\"30234\" data-gtm-vis-total-visible-time48430672_468=\"1000\" data-gtm-vis-first-on-screen48430672_493=\"30235\" data-gtm-vis-total-visible-time48430672_493=\"1000\" data-gtm-vis-first-on-screen48430672_501=\"30235\" data-gtm-vis-total-visible-time48430672_501=\"1000\" data-gtm-vis-first-on-screen48430672_505=\"30237\" data-gtm-vis-total-visible-time48430672_505=\"1000\" data-gtm-vis-first-on-screen48430672_506=\"30238\" data-gtm-vis-total-visible-time48430672_506=\"1000\" data-gtm-vis-first-on-screen48430672_531=\"30238\" data-gtm-vis-total-visible-time48430672_531=\"1000\" data-gtm-vis-has-fired48430672_37=\"1\" data-gtm-vis-has-fired48430672_468=\"1\" data-gtm-vis-has-fired48430672_493=\"1\" data-gtm-vis-has-fired48430672_501=\"1\" data-gtm-vis-has-fired48430672_505=\"1\" data-gtm-vis-has-fired48430672_506=\"1\" data-gtm-vis-has-fired48430672_531=\"1\">\n<div class=\"col-1-13 [&amp;_*]:font-serif [&amp;_&gt;*:last-child]:mb-0 [&amp;_h2]:mt-8 [&amp;_h2]:mb-4 [&amp;_h2]:h-editorial-030 [&amp;_p]:break-words [&amp;_p]:my-4 [&amp;_p]:text-p sm:[&amp;_p]:my-6 [&amp;_ul]:my-4 [&amp;_ul]:text-p sm:[&amp;_ul]:my-6 [&amp;_ol]:my-4 [&amp;_ol]:text-p sm:[&amp;_ol]:my-6 [&amp;_p:last-of-type]:pb-0 [&amp;_p:last-of-type]:mb-0 [&amp;_a]:underline [&amp;_a]:text-interactive-tertiary [&amp;_a:active]:no-underline [&amp;_a:hover]:no-underline [&amp;_a:visited]:text-interactive-tertiary [&amp;_figure]:my-6 sm:[&amp;_figure]:my-10 [&amp;_figure_figcaption]:mt-2 [&amp;_figure_figcaption]:font-sans [&amp;_figure_figcaption]:leading-6 [&amp;_figure_figcaption]:text-caption sm:[&amp;_figure_figcaption]:mt-3 [&amp;_figure_figcaption],[&amp;_figure_figcaption_a],[&amp;_figure_figcaption_a:visited]:[&amp;_figure_figcaption]:text-soft-black-shade [&amp;_figure_figcaption_a],[&amp;_figure_figcaption_a:visited]:font-sans [&amp;_figure_figcaption_a]:underline [&amp;_figure&gt;img]:w-full [&amp;_blockquote:not(.instagram-media)]:pb-2 [&amp;_blockquote:not(.instagram-media)]:pl-8 [&amp;_blockquote:not(.instagram-media)]:mt-6 [&amp;_blockquote:not(.instagram-media)]:mb-0 [&amp;_blockquote:not(.instagram-media)]:italic [&amp;_blockquote:not(.instagram-media)]:border-l [&amp;_blockquote:not(.instagram-media)]:border-solid [&amp;_blockquote:not(.instagram-media)]:border-accent-blue-tint sm:[&amp;_blockquote:not(.instagram-media)]:mt-10 sm:[&amp;_blockquote:not(.instagram-media)]:pb-6 [&amp;_blockquote:not(.instagram-media)]:text-p [&amp;_blockquote:not(.instagram-media)]:margin-inline-start-0 [&amp;_blockquote:not(.instagram-media)]:margin-inline-end-0 [&amp;_blockquote+*]:mt-6 sm:[&amp;_blockquote+*]:mt-10 [&amp;_blockquote+cite]:inline-block [&amp;_blockquote+cite]:pt-0 [&amp;_blockquote+cite]:mt-0 [&amp;_blockquote+cite]:text-p [&amp;_blockquote+cite::before]:text-base sm:[&amp;_blockquote+cite::before]:text-caption [&amp;_blockquote+cite::before]:content-[&quot;&quot;] [&amp;_blockquote+blockquote]:pt-8 sm:[&amp;_blockquote+blockquote]:pt-6 [&amp;_blockquote+blockquote]:-mt-8 sm:[&amp;_blockquote+blockquote]:-mt-12 [&amp;_cite]:pl-8 [&amp;_cite]:border-l [&amp;_cite]:border-solid [&amp;_cite]:text-soft-black-shade [&amp;_cite]:border-accent-blue-tint [&amp;_cite]:text-p [&amp;_cite+*]:mt-6 sm:[&amp;_cite+*]:mt-10 [&amp;_strong]:font-bold [&amp;_u]:no-underline [&amp;_ol]:pl-0 [&amp;_ol]:list-decimal [&amp;_ol]:list-inside [&amp;_ul]:list-image-bullet-filled [&amp;_ul]:pt-0 [&amp;_ul]:pr-0 [&amp;_ul]:pb-0 [&amp;_ul]:pl-4.5 [&amp;_ul_li]:pl-2 [&amp;_ul_li:not(:last-of-type)]:mb-3 [&amp;_ul_ul]:list-image-bullet-outlined [&amp;_ul_li]:mt-4 sm:[&amp;_ul_li]:mt-6 [&amp;_ol_li]:mt-4 sm:[&amp;_ol_li]:mt-6 [&amp;_.iframe-responsive-wrapper]:pb-[56.25%] [&amp;_.iframe-responsive-wrapper]:relative [&amp;_.iframe-responsive-wrapper]:overflow-hidden [&amp;_.iframe-responsive-wrapper_iframe]:h-full [&amp;_.iframe-responsive-wrapper_iframe]:left-0 [&amp;_.iframe-responsive-wrapper_iframe]:absolute [&amp;_.iframe-responsive-wrapper_iframe]:top-0 [&amp;_.iframe-responsive-wrapper_iframe]:w-full [&amp;_.iframe-responsive-wrapper:has(&gt;iframe[src*=&quot;documentcloud.org&quot;])]:pb-[max(56.25%,500px)] [&amp;_iframe[src*=&quot;documentcloud.org&quot;]]:border [&amp;_iframe[src*=&quot;documentcloud.org&quot;]]:border-[#aaa] [&amp;_iframe[id*=&quot;datawrapper-chart&quot;]]:border-none [&amp;_iframe[id*=&quot;datawrapper-chart&quot;]]:min-w-full [&amp;_iframe[id*=&quot;datawrapper-chart&quot;]]:w-0 xl:[&amp;_iframe[id*=&quot;datawrapper-chart&quot;]]:min-h-[465px] [&amp;_img[src$=&quot;giphy.gif&quot;]]:mx-auto [&amp;_.twitter-tweet]:mx-auto [&amp;_.instagram-wrapper]:flex [&amp;_.instagram-wrapper]:justify-center [&amp;_.axios-visual-newsletter-fallback-image]:hidden sm:[&amp;_.axios-visual-newsletter-fallback-image]:block sm:[&amp;_.axios-visual-apple-fallback-image]:hidden [&amp;_div.subhead-timestamp]:-mt-15 [&amp;_div.subhead-timestamp]:pt-15 sm:[&amp;_div.subhead-timestamp]:-mt-20 sm:[&amp;_div.subhead-timestamp]:pt-20 [&amp;_div.subhead-timestamp_*]:font-sans [&amp;_div.subhead-timestamp_div]:flex [&amp;_div.subhead-timestamp_div]:gap-1 [&amp;_div.subhead-timestamp_a]:no-underline [&amp;_div.subhead-timestamp_a]:text-soft-black-shade [&amp;_div.subhead-timestamp_a_h2]:mt-0 &amp;.story-is-live [&amp;_div.subhead-timestamp_div_time]:text-red-core [&amp;_#piano-wall:has(#piano-container[data-piano-active])~.gated-content]:hidden [&amp;_#piano-container:not(:empty)]:relative [&amp;_#piano-container:not(:empty)]:before:content-[&quot;&quot;] [&amp;_#piano-container:not(:empty)]:before:h-40 [&amp;_#piano-container:not(:empty)]:before:absolute [&amp;_#piano-container:not(:empty)]:before:-top-40 [&amp;_#piano-container:not(:empty)]:before:w-full [&amp;_#piano-container:not(:empty)]:before:bg-gradient-to-b [&amp;_#piano-container:not(:empty)]:before:from-transparent [&amp;_#piano-container:not(:empty)]:before:to-white [&amp;_#piano-container:not(:empty)]:before:pointer-events-none [&amp;_#piano-spinner]:relative [&amp;_#piano-spinner]:before:content-[&quot;&quot;] [&amp;_#piano-spinner]:before:h-40 [&amp;_#piano-spinner]:before:absolute [&amp;_#piano-spinner]:before:-top-40 [&amp;_#piano-spinner]:before:w-full [&amp;_#piano-spinner]:before:bg-gradient-to-b [&amp;_#piano-spinner]:before:from-transparent [&amp;_#piano-spinner]:before:to-white [&amp;_#piano-spinner]:before:pointer-events-none [&amp;_#paywall-fallback-container]:relative [&amp;_#paywall-fallback-container]:before:content-[&quot;&quot;] [&amp;_#paywall-fallback-container]:before:h-40 [&amp;_#paywall-fallback-container]:before:absolute [&amp;_#paywall-fallback-container]:before:-top-40 [&amp;_#paywall-fallback-container]:before:w-full [&amp;_#paywall-fallback-container]:before:bg-gradient-to-b [&amp;_#paywall-fallback-container]:before:from-transparent [&amp;_#paywall-fallback-container]:before:to-white [&amp;_#paywall-fallback-container]:before:pointer-events-none [&amp;_figure&gt;figcaption&gt;span]:font-sans\" data-chromatic=\"ignore\">\n<p>OpenAI says the cyber capabilities of its frontier\u00a0<a class=\"gtmContentClick\" href=\"https:\/\/www.axios.com\/technology\/automation-and-ai\" target=\"_self\" data-vars-link-text=\"AI\" data-vars-click-url=\"https:\/\/www.axios.com\/technology\/automation-and-ai\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-event-category=\"story\" data-vars-sub-category=\"story\" data-vars-item=\"in_content_link\">AI<\/a>\u00a0models are accelerating and warns Wednesday that upcoming models are likely to pose a &#8220;high&#8221; risk, according to a report shared first with Axios.<\/p>\n<p><strong>Why it matters:\u00a0<\/strong>The models&#8217; growing capabilities could significantly expand the number of people able to carry out cyberattacks.<\/p>\n<p><strong>Driving the news:\u00a0<\/strong>OpenAI said it has already seen a significant increase in capabilities in recent releases, particularly as models are able to operate longer autonomously, paving the way for brute force attacks.<\/p>\n<ul>\n<li>The company notes that GPT-5 scored a 27% on a capture-the-flag exercise in August, GPT-5.1-Codex-Max was able to score 76% last month.<\/li>\n<li>&#8220;We expect that upcoming AI models will continue on this trajectory,&#8221; the company says in the report. &#8220;In preparation, we are planning and evaluating as though each new model could reach &#8216;high&#8217; levels of cybersecurity capability as measured by our\u00a0<a class=\"gtmContentClick\" href=\"https:\/\/www.axios.com\/2023\/12\/18\/openai-ai-safety-risk-catastrophic-preparedness\" target=\"_self\" data-vars-link-text=\"Preparedness Framework\" data-vars-click-url=\"https:\/\/www.axios.com\/2023\/12\/18\/openai-ai-safety-risk-catastrophic-preparedness\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-event-category=\"story\" data-vars-sub-category=\"story\" data-vars-item=\"in_content_link\">Preparedness Framework<\/a>.&#8221;<\/li>\n<\/ul>\n<p><strong>Catch up quick<\/strong>: OpenAI issued a\u00a0<a class=\"gtmContentClick\" href=\"https:\/\/www.axios.com\/2025\/06\/18\/openai-bioweapons-risk\" target=\"_self\" data-vars-link-text=\"similar warning\" data-vars-click-url=\"https:\/\/www.axios.com\/2025\/06\/18\/openai-bioweapons-risk\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-event-category=\"story\" data-vars-sub-category=\"story\" data-vars-item=\"in_content_link\">similar warning<\/a>\u00a0relative to bioweapons risk in June, and then released ChatGPT Agent in July, which\u00a0<a class=\"gtmContentClick\" href=\"https:\/\/fortune.com\/2025\/07\/18\/openai-chatgpt-agent-could-aid-dangerous-bioweapon-development\/\" target=\"_blank\" rel=\"noopener\" data-vars-link-text=\"did indeed\" data-vars-click-url=\"https:\/\/fortune.com\/2025\/07\/18\/openai-chatgpt-agent-could-aid-dangerous-bioweapon-development\/\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-event-category=\"story\" data-vars-sub-category=\"story\" data-vars-item=\"in_content_link\">did indeed<\/a>\u00a0rate &#8220;high&#8221; on its risk levels.<\/p>\n<ul>\n<li>&#8220;High&#8221; is the second-highest level, below the &#8220;critical&#8221; level at which models are unsafe to be released publicly.<\/li>\n<\/ul>\n<p><strong>Yes, but:<\/strong>\u00a0The company didn&#8217;t say exactly when to expect the first models rated &#8220;high&#8221; for cybersecurity risk, or which types of future models could pose such a risk.<\/p>\n<p><strong>What they&#8217;re saying:\u00a0<\/strong>&#8220;What I would explicitly call out as the forcing function for this is the model&#8217;s ability to work for extended periods of time,&#8221; OpenAI&#8217;s Fouad Matin told Axios in an exclusive interview.<\/p>\n<ul>\n<li>These kinds of brute force attacks that rely on this extended time are more easily defended, Matin says.<\/li>\n<\/ul>\n<div class=\"gtmView\" data-vars-experiment=\"mid_story_ad\" data-vars-experiment-variant=\"ad\" data-vars-sub-category=\"non-braze\" data-vars-product=\"core\" data-vars-page-type=\"story\" data-vars-sub-product=\"national\" data-vars-latitude=\"55.95\" data-vars-longitude=\"-3.2\" data-vars-postal-code=\"EH1\" data-vars-event-name=\"cta_view\" data-vars-deprecated-category=\"cta\" data-vars-deprecated-experiment=\"mid_story_ad\" data-vars-deprecated-experiment-variant=\"ad\" data-vars-deprecated-label=\"cta\/view\/mid_story_ad\/ad\" data-gtm-vis-first-on-screen48430672_506=\"46499\" data-gtm-vis-total-visible-time48430672_506=\"100\" data-gtm-vis-first-on-screen48430672_37=\"46499\" data-gtm-vis-total-visible-time48430672_37=\"100\" data-gtm-vis-first-on-screen48430672_531=\"46499\" data-gtm-vis-total-visible-time48430672_531=\"100\" data-gtm-vis-first-on-screen48430672_468=\"46499\" data-gtm-vis-total-visible-time48430672_468=\"100\" data-gtm-vis-first-on-screen48430672_505=\"46500\" data-gtm-vis-total-visible-time48430672_505=\"100\" data-gtm-vis-first-on-screen48430672_493=\"46504\" data-gtm-vis-total-visible-time48430672_493=\"100\" data-gtm-vis-first-on-screen48430672_501=\"46504\" data-gtm-vis-total-visible-time48430672_501=\"100\">\n<div class=\"mt-6 mb-6\">\n<div class=\"gtmView midStoryAd grid justify-center items-center py-2 overflow-hidden lg:overflow-visible\" data-cy=\"mid-story-ad\" data-vars-event-name=\"ad_view\" data-vars-product=\"core\" data-vars-page-type=\"story\" data-vars-sub-product=\"national\" data-vars-latitude=\"55.95\" data-vars-longitude=\"-3.2\" data-vars-postal-code=\"EH1\" data-vars-sub-category=\"mid_story\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-gtm-vis-first-on-screen48430672_506=\"46499\" data-gtm-vis-total-visible-time48430672_506=\"100\" data-gtm-vis-first-on-screen48430672_37=\"46499\" data-gtm-vis-total-visible-time48430672_37=\"100\" data-gtm-vis-first-on-screen48430672_531=\"46499\" data-gtm-vis-total-visible-time48430672_531=\"100\" data-gtm-vis-first-on-screen48430672_468=\"46499\" data-gtm-vis-total-visible-time48430672_468=\"100\" data-gtm-vis-first-on-screen48430672_505=\"46504\" data-gtm-vis-total-visible-time48430672_505=\"100\" data-gtm-vis-first-on-screen48430672_493=\"46504\" data-gtm-vis-total-visible-time48430672_493=\"100\" data-gtm-vis-first-on-screen48430672_501=\"46504\" data-gtm-vis-total-visible-time48430672_501=\"100\">\n<div class=\"adunitContainer\"><\/div>\n<\/div>\n<\/div>\n<\/div>\n<ul>\n<li>&#8220;In any defended environment this would be caught pretty easily,&#8221; he added.<\/li>\n<\/ul>\n<p><strong>The big picture:\u00a0<\/strong>Leading models are getting better at finding security vulnerabilities \u2014 and not just models from OpenAI.<\/p>\n<ul>\n<li>As a result, OpenAI says it has been stepping up efforts to work across the industry on cybersecurity threats, including through the\u00a0<a class=\"gtmContentClick\" href=\"https:\/\/www.axios.com\/2023\/07\/26\/ai-frontier-model-forum-established\" target=\"_self\" data-vars-link-text=\"Frontier Model Forum\" data-vars-click-url=\"https:\/\/www.axios.com\/2023\/07\/26\/ai-frontier-model-forum-established\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-event-category=\"story\" data-vars-sub-category=\"story\" data-vars-item=\"in_content_link\">Frontier Model Forum<\/a>\u00a0that it started with other leading labs in 2023.<\/li>\n<li>The company says it will establish a separate Frontier Risk Council, an advisory group that will &#8220;bring experienced cyber defenders and security practitioners into close collaboration&#8221; with OpenAI&#8217;s teams.<\/li>\n<li>It&#8217;s also in private testing for Aardvark, a tool that developers can use to find security gaps in their products. Developers have to\u00a0<a class=\"gtmContentClick\" href=\"https:\/\/openai.com\/form\/aardvark-beta-signup\/\" target=\"_blank\" rel=\"noopener\" data-vars-link-text=\"apply\" data-vars-click-url=\"https:\/\/openai.com\/form\/aardvark-beta-signup\/\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-event-category=\"story\" data-vars-sub-category=\"story\" data-vars-item=\"in_content_link\">apply<\/a>\u00a0to gain access to Aardvark, which has already found critical vulnerabilities, OpenAI said.<\/li>\n<\/ul>\n<\/div>\n<\/div>\n<div class=\"flex flex-col gap-4 xs:mb-6 sm:mb-10 md:flex-row\">\n<div class=\"gtmView\" data-vars-event-name=\"share_view\" data-vars-content-id=\"f39896e3-8347-4dbb-b56b-71c89ac0ef4c\" data-vars-headline=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says\" data-vars-location=\"bottom\" data-vars-latitude=\"55.95\" data-vars-longitude=\"-3.2\" data-vars-postal-code=\"EH1\" data-vars-product=\"core\" data-vars-page-type=\"story\" data-vars-sub-product=\"national\" data-gtm-vis-recent-on-screen48430672_506=\"46663\" data-gtm-vis-first-on-screen48430672_506=\"46663\" data-gtm-vis-total-visible-time48430672_506=\"1000\" data-gtm-vis-recent-on-screen48430672_37=\"46670\" data-gtm-vis-first-on-screen48430672_37=\"46670\" data-gtm-vis-total-visible-time48430672_37=\"1000\" data-gtm-vis-recent-on-screen48430672_531=\"46670\" data-gtm-vis-first-on-screen48430672_531=\"46670\" data-gtm-vis-total-visible-time48430672_531=\"1000\" data-gtm-vis-recent-on-screen48430672_468=\"46673\" data-gtm-vis-first-on-screen48430672_468=\"46673\" data-gtm-vis-total-visible-time48430672_468=\"1000\" data-gtm-vis-recent-on-screen48430672_505=\"46673\" data-gtm-vis-first-on-screen48430672_505=\"46674\" data-gtm-vis-total-visible-time48430672_505=\"1000\" data-gtm-vis-recent-on-screen48430672_493=\"46674\" data-gtm-vis-first-on-screen48430672_493=\"46674\" data-gtm-vis-total-visible-time48430672_493=\"1000\" data-gtm-vis-recent-on-screen48430672_501=\"46674\" data-gtm-vis-first-on-screen48430672_501=\"46674\" data-gtm-vis-total-visible-time48430672_501=\"1000\" data-gtm-vis-has-fired48430672_37=\"1\" data-gtm-vis-has-fired48430672_468=\"1\" data-gtm-vis-has-fired48430672_493=\"1\" data-gtm-vis-has-fired48430672_501=\"1\" data-gtm-vis-has-fired48430672_505=\"1\" data-gtm-vis-has-fired48430672_506=\"1\" data-gtm-vis-has-fired48430672_531=\"1\">\n<ul class=\"space-x-4 leading-0\" data-cy=\"social-share-bottom\">\n<li class=\"inline\">\n<div class=\"group\"><span class=\"fixed sr-only\">facebook (opens in new window)<\/span><\/div>\n<\/li>\n<li class=\"inline\">\n<div class=\"group\"><span class=\"fixed sr-only\">twitter (opens in new window)<\/span><\/div>\n<\/li>\n<li class=\"inline\">\n<div class=\"group\"><span class=\"fixed sr-only\">linkedin (opens in new window)<\/span><\/div>\n<\/li>\n<li class=\"inline\">\n<div class=\"group\"><span class=\"fixed sr-only\">email (opens in new window)<\/span><\/div>\n<\/li>\n<\/ul>\n<p><a href=\"https:\/\/www.axios.com\/2025\/12\/10\/openai-new-models-cybersecurity-risks\">Read the Full Story Here<\/a><\/p>\n<p>Source: Axios<\/p>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>OpenAI says the cyber capabilities of its frontier\u00a0AI\u00a0models are accelerating and warns Wednesday that upcoming models are likely to pose a &#8220;high&#8221; risk, according to a report shared first with Axios. Why it matters:\u00a0The models&#8217; growing capabilities could significantly expand the number of people able to carry out cyberattacks. Driving the news:\u00a0OpenAI said it has [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":1529,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[4],"tags":[],"class_list":["post-1835","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v24.9 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says - Community<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Exclusive: New OpenAI models likely pose &quot;high&quot; cybersecurity risk, company says - Community\" \/>\n<meta property=\"og:description\" content=\"OpenAI says the cyber capabilities of its frontier\u00a0AI\u00a0models are accelerating and warns Wednesday that upcoming models are likely to pose a &#8220;high&#8221; risk, according to a report shared first with Axios. Why it matters:\u00a0The models&#8217; growing capabilities could significantly expand the number of people able to carry out cyberattacks. Driving the news:\u00a0OpenAI said it has [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/\" \/>\n<meta property=\"og:site_name\" content=\"Community\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/TSFactoryLLC\/\" \/>\n<meta property=\"article:published_time\" content=\"2025-12-11T10:42:29+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-12-11T10:44:14+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1280\" \/>\n\t<meta property=\"og:image:height\" content=\"1920\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Chelsie Wyatt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@TSFactoryLLC\" \/>\n<meta name=\"twitter:site\" content=\"@TSFactoryLLC\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Chelsie Wyatt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/\",\"url\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/\",\"name\":\"Exclusive: New OpenAI models likely pose \\\"high\\\" cybersecurity risk, company says - Community\",\"isPartOf\":{\"@id\":\"https:\/\/www.tsfactory.com\/forums\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg\",\"datePublished\":\"2025-12-11T10:42:29+00:00\",\"dateModified\":\"2025-12-11T10:44:14+00:00\",\"author\":{\"@id\":\"https:\/\/www.tsfactory.com\/forums\/#\/schema\/person\/9d9908f0e12559297335ebe9b601c82f\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#primaryimage\",\"url\":\"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg\",\"contentUrl\":\"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg\",\"width\":1280,\"height\":1920},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.tsfactory.com\/forums\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Exclusive: New OpenAI models likely pose &#8220;high&#8221; cybersecurity risk, company says\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.tsfactory.com\/forums\/#website\",\"url\":\"https:\/\/www.tsfactory.com\/forums\/\",\"name\":\"Community\",\"description\":\"TSFactory\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.tsfactory.com\/forums\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.tsfactory.com\/forums\/#\/schema\/person\/9d9908f0e12559297335ebe9b601c82f\",\"name\":\"Chelsie Wyatt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.tsfactory.com\/forums\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/09ff3801fb7566acf715fe4e81a9bd942b923c236138a3ed8a8375f099e5d6d6?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/09ff3801fb7566acf715fe4e81a9bd942b923c236138a3ed8a8375f099e5d6d6?s=96&d=mm&r=g\",\"caption\":\"Chelsie Wyatt\"},\"url\":\"https:\/\/www.tsfactory.com\/forums\/blog\/author\/chelsie\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Exclusive: New OpenAI models likely pose \"high\" cybersecurity risk, company says - Community","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/","og_locale":"en_US","og_type":"article","og_title":"Exclusive: New OpenAI models likely pose \"high\" cybersecurity risk, company says - Community","og_description":"OpenAI says the cyber capabilities of its frontier\u00a0AI\u00a0models are accelerating and warns Wednesday that upcoming models are likely to pose a &#8220;high&#8221; risk, according to a report shared first with Axios. Why it matters:\u00a0The models&#8217; growing capabilities could significantly expand the number of people able to carry out cyberattacks. Driving the news:\u00a0OpenAI said it has [&hellip;]","og_url":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/","og_site_name":"Community","article_publisher":"https:\/\/www.facebook.com\/TSFactoryLLC\/","article_published_time":"2025-12-11T10:42:29+00:00","article_modified_time":"2025-12-11T10:44:14+00:00","og_image":[{"width":1280,"height":1920,"url":"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg","type":"image\/jpeg"}],"author":"Chelsie Wyatt","twitter_card":"summary_large_image","twitter_creator":"@TSFactoryLLC","twitter_site":"@TSFactoryLLC","twitter_misc":{"Written by":"Chelsie Wyatt","Est. reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/","url":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/","name":"Exclusive: New OpenAI models likely pose \"high\" cybersecurity risk, company says - Community","isPartOf":{"@id":"https:\/\/www.tsfactory.com\/forums\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#primaryimage"},"image":{"@id":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#primaryimage"},"thumbnailUrl":"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg","datePublished":"2025-12-11T10:42:29+00:00","dateModified":"2025-12-11T10:44:14+00:00","author":{"@id":"https:\/\/www.tsfactory.com\/forums\/#\/schema\/person\/9d9908f0e12559297335ebe9b601c82f"},"breadcrumb":{"@id":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#primaryimage","url":"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg","contentUrl":"https:\/\/www.tsfactory.com\/forums\/wp-content\/uploads\/2025\/01\/openai.jpg","width":1280,"height":1920},{"@type":"BreadcrumbList","@id":"https:\/\/www.tsfactory.com\/forums\/blog\/exclusive-new-openai-models-likely-pose-high-cybersecurity-risk-company-says\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.tsfactory.com\/forums\/"},{"@type":"ListItem","position":2,"name":"Exclusive: New OpenAI models likely pose &#8220;high&#8221; cybersecurity risk, company says"}]},{"@type":"WebSite","@id":"https:\/\/www.tsfactory.com\/forums\/#website","url":"https:\/\/www.tsfactory.com\/forums\/","name":"Community","description":"TSFactory","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.tsfactory.com\/forums\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/www.tsfactory.com\/forums\/#\/schema\/person\/9d9908f0e12559297335ebe9b601c82f","name":"Chelsie Wyatt","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.tsfactory.com\/forums\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/09ff3801fb7566acf715fe4e81a9bd942b923c236138a3ed8a8375f099e5d6d6?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/09ff3801fb7566acf715fe4e81a9bd942b923c236138a3ed8a8375f099e5d6d6?s=96&d=mm&r=g","caption":"Chelsie Wyatt"},"url":"https:\/\/www.tsfactory.com\/forums\/blog\/author\/chelsie\/"}]}},"_links":{"self":[{"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/posts\/1835","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/comments?post=1835"}],"version-history":[{"count":2,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/posts\/1835\/revisions"}],"predecessor-version":[{"id":1837,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/posts\/1835\/revisions\/1837"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/media\/1529"}],"wp:attachment":[{"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/media?parent=1835"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/categories?post=1835"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.tsfactory.com\/forums\/wp-json\/wp\/v2\/tags?post=1835"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}