{"id":85763,"date":"2026-08-25T10:08:44","date_gmt":"2026-08-25T14:08:44","guid":{"rendered":"https:\/\/spiritdaily.org\/blog\/?p=85763"},"modified":"2026-08-25T10:08:44","modified_gmt":"2026-08-25T14:08:44","slug":"a-i-in-sobering-hacks","status":"publish","type":"post","link":"https:\/\/spiritdaily.org\/blog\/a-i-in-sobering-hacks\/","title":{"rendered":"A.I. In Sobering Hacks"},"content":{"rendered":"<p><em>From The New York Times:<\/em><\/p>\n<p class=\"css-12m5bll evys1bk0\">OpenAI recently discovered that a new artificial intelligence model it was testing had gone rogue and\u00a0<a class=\"css-povzk\" title=\"\" href=\"https:\/\/www.nytimes.com\/2026\/07\/21\/technology\/openai-attack-hugging-face.html\">hacked another company<\/a>.\u00a0Anthropic then revealed that one of its A.I. models had\u00a0<a class=\"css-povzk\" title=\"\" href=\"https:\/\/www.nytimes.com\/2026\/07\/30\/technology\/anthropic-ai-hack.html\">broken into the systems of three outside organizations<\/a> during a test.\u00a0Not long after, Meta said its A.I. models had done something similar.<\/p>\n<p class=\"css-12m5bll evys1bk0\">All three incidents had one company in common: Irregular, an Israeli start-up that works with the Silicon Valley giants to assess their A.I. models before the technology is publicly released. The firm \u2014 which conducted the tests that went awry \u2014 is part of a group of start-ups that are doing the novel work of scrutinizing cutting-edge A.I. models to gauge their sophistication and check their security. The goal is to instill public confidence in the models and to prevent them from being misused.\u00a0The recent breaches occurred when\u00a0<a class=\"css-povzk\" title=\"\" href=\"https:\/\/www.irregular.com\/research\/addressing-recent-incidents-ongoing-findings-and-path-forward\" target=\"_blank\" rel=\"noopener noreferrer\">Irregular made an error<\/a>\u00a0during the tests with the models from Anthropic, OpenAI and Meta. But the A.I. models then compounded the situations by acting in powerful and unexpected ways, said Dan Lahav, the chief executive of Irregular.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>From The New York Times: OpenAI recently discovered that a new artificial intelligence model it was testing had gone rogue.<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_genesis_hide_title":false,"_genesis_hide_breadcrumbs":false,"_genesis_hide_singular_image":false,"_genesis_hide_footer_widgets":false,"_genesis_custom_body_class":"","_genesis_custom_post_class":"","_genesis_layout":"","footnotes":""},"categories":[32],"tags":[],"class_list":{"0":"post-85763","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-news","7":"entry"},"acf":[],"_links":{"self":[{"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/posts\/85763","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/comments?post=85763"}],"version-history":[{"count":1,"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/posts\/85763\/revisions"}],"predecessor-version":[{"id":85764,"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/posts\/85763\/revisions\/85764"}],"wp:attachment":[{"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/media?parent=85763"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/categories?post=85763"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/spiritdaily.org\/blog\/wp-json\/wp\/v2\/tags?post=85763"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}