{"id":28387,"date":"2026-05-05T18:50:27","date_gmt":"2026-05-05T18:50:27","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/28387\/"},"modified":"2026-05-05T18:50:27","modified_gmt":"2026-05-05T18:50:27","slug":"microsoft-xai-and-google-will-share-ai-models-with-us-government-for-security-reviews-2","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/28387\/","title":{"rendered":"Microsoft, xAI and Google will share AI models with US government for security reviews"},"content":{"rendered":"<p><a href=\"https:\/\/economictimes.indiatimes.com\/topic\/microsoft\" target=\"_blank\" rel=\"nofollow noopener\">Microsoft<\/a>, Alphabet-owned <a href=\"https:\/\/economictimes.indiatimes.com\/topic\/google\" target=\"_blank\" rel=\"nofollow noopener\">Google<\/a> and Elon Musk&#8217;s <a href=\"https:\/\/economictimes.indiatimes.com\/topic\/xai\" target=\"_blank\" rel=\"nofollow noopener\">xAI<\/a> will give the U.S. government early access to new artificial intelligence models before their public release to allow checks for <a href=\"https:\/\/economictimes.indiatimes.com\/topic\/national-security\" target=\"_blank\" rel=\"nofollow noopener\">national security<\/a> risks under a new deal.<br \/>The Center for AI Standards and Innovation (<a href=\"https:\/\/economictimes.indiatimes.com\/topic\/caisi\" target=\"_blank\" rel=\"nofollow noopener\">CAISI<\/a>) at the Department of Commerce said on Tuesday that the agreement would allow \u200cit to evaluate \u2060the \u2060models before deployment and conduct research to assess their capabilities and security risks.<\/p>\n<p>The development of advanced AI systems including <a href=\"https:\/\/economictimes.indiatimes.com\/topic\/anthropic\" target=\"_blank\" rel=\"nofollow noopener\">Anthropic<\/a>&#8216;s Mythos has in recent weeks created a stir globally, including among U.S. officials and corporate America, over their ability to supercharge hackers.<\/p>\n<p>The ChatGPT-maker unveiled <a href=\"https:\/\/economictimes.indiatimes.com\/topic\/gpt-5-4-cyber\" target=\"_blank\" rel=\"nofollow noopener\">GPT-5.4-Cyber<\/a> last month, a variant of its latest flagship model fine-tuned specifically for defensive cybersecurity work, following \u2060rival Anthropic&#8217;s \u200cannouncement of advanced AI model Mythos.<\/p>\n<p>Anthropic has been embroiled \u200bin a \u200bdispute with the Pentagon over guardrails on the military&#8217;s \u2060use of its AI tools.<\/p>\n<p>&#8220;Independent, rigorous measurement science is \u200bessential to understanding frontier AI and its national security \u200bimplications,&#8221; CAISI Director Chris Fall said in a statement.<\/p>\n<p>The move builds on 2024 agreements with OpenAI and Anthropic, established under the Biden administration when CAISI was known as the U.S. Artificial Intelligence Safety Institute.<\/p>\n<p>CAISI, which serves as the government&#8217;s main hub for AI model testing, said it had \u200calready completed more than 40 evaluations, including on cutting-edge models not yet available to the public.<\/p>\n<p>Developers frequently hand over versions \u200bof their \u200bmodels with safety guardrails \u2060stripped back so the center can probe for national security risks, the agency said.<\/p>\n<p>Last week, the Pentagon said it had reached agreements with seven AI \u200bcompanies to deploy their advanced capabilities on the Defense Department&#8217;s classified networks as it seeks to broaden the range of AI providers working across the military.<\/p>\n<p>Microsoft, Google and xAI did not immediately respond to requests for comment.<\/p>\n","protected":false},"excerpt":{"rendered":"Microsoft, Alphabet-owned Google and Elon Musk&#8217;s xAI will give the U.S. government early access to new artificial intelligence&hellip;\n","protected":false},"author":2,"featured_media":28388,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[1276,53,18719,18501,132,7640,320,1896,18720,2899],"class_list":{"0":"post-28387","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-xai","8":"tag-ai-models","9":"tag-anthropic","10":"tag-artificial-intelligence-security","11":"tag-caisi","12":"tag-google","13":"tag-gpt-5-4-cyber","14":"tag-microsoft","15":"tag-national-security","16":"tag-pentagon-ai-agreements","17":"tag-xai"},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/28387","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=28387"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/28387\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/28388"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=28387"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=28387"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=28387"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}