{"id":639576,"date":"2026-08-06T04:52:40","date_gmt":"2026-08-06T04:52:40","guid":{"rendered":"https:\/\/e-bitco.in\/index.php\/2026\/08\/06\/meta-latest-ai-firm-to-see-model-go-rogue-during-testing\/"},"modified":"2026-08-06T04:52:40","modified_gmt":"2026-08-06T04:52:40","slug":"meta-latest-ai-firm-to-see-model-go-rogue-during-testing","status":"publish","type":"post","link":"https:\/\/e-bitco.in\/index.php\/2026\/08\/06\/meta-latest-ai-firm-to-see-model-go-rogue-during-testing\/","title":{"rendered":"Meta latest AI firm to see model go rogue during testing"},"content":{"rendered":"<p><img decoding=\"async\" src=\"https:\/\/s3-images.ctmedia.io\/media\/article-covers\/ai.jpg\" alt=\"Meta latest AI firm to see model go rogue during testing\" class=\"type:primaryImage\" \/><\/p>\n<p>The incident reportedly stemmed from a misconfigured testing environment, adding Meta to a growing list of AI firms whose models have escaped evaluation sandboxes. <\/p>\n<p>Meta has become the latest major AI company to disclose that one of its models hacked another company\u2019s systems during testing, following similar incidents involving Anthropic and OpenAI.\u00a0<\/p>\n<p>The model involved Meta\u2019s Muse Spark 1.1, which launched in July, according to The Information, <a href=\"https:\/\/www.theinformation.com\/articles\/meta-ai-model-hacked-another-company-cybersecurity-testing\"><span>citing<\/span><\/a> sources. The issue reportedly stemmed from a misconfiguration by Irregular, an artificial intelligence security testing and red-teaming firm, which inadvertently gave the model internet access during an evaluation.\u00a0\u00a0<\/p>\n<p>The model \u201cexploited a security vulnerability in a third-party service, in a manner similar to previously reported instances with other companies,\u201d Meta told Reuters in a statement.\u00a0<\/p>\n<p><a href=\"https:\/\/cointelegraph.com\/news\/meta-latest-ai-firm-to-see-model-go-rogue-during-testing\">Read more<\/a><\/p>","protected":false},"excerpt":{"rendered":"<p>The incident reportedly stemmed from a misconfigured testing environment, adding Meta to a growing list of AI firms whose models have escaped evaluation sandboxes. Meta has become the latest major AI company to disclose that one of its models hacked another company\u2019s systems during testing, following similar incidents involving Anthropic and OpenAI.\u00a0 The model involved [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":639577,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[12],"tags":[],"class_list":{"0":"post-639576","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-blockchain"},"_links":{"self":[{"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/posts\/639576","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/comments?post=639576"}],"version-history":[{"count":0,"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/posts\/639576\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/media\/639577"}],"wp:attachment":[{"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/media?parent=639576"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/categories?post=639576"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/e-bitco.in\/index.php\/wp-json\/wp\/v2\/tags?post=639576"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}