Anthropic said the review did not find any \u201cincidents of similar severity to the cybersecurity incidents we reported this summer.\u201d The review later moved on to transcripts of \u201clower-risk\u201d testing scenarios like the one affecting the Philadelphia Police Department.<\/p>\n
\u201cAlthough the impact of these behaviors was minimal and we had already turned off live Internet access for some high-risk and cybersecurity evaluations, we have now decided to expand that to include all our internal evaluations until we have confirmed that our security and monitoring measures\u2026 reliably catch behaviors like these,\u201d Anthropic said.<\/p>\n
Anthropic says new tools block unintended behaviors<\/p>\n
Anthropic said it has stopped running some evaluations and moved others to offline versions \u201cor rebuilt them so that their tasks do not reach live websites.\u201d The company \u201cupdated the guardrails on some of our Internet access tools, such as the web fetch tool, to heavily restrict what the model can do with them,\u201d it said.<\/p>\n
Anthropic said it also built new tools \u201cto automatically detect and block the kinds of behaviors described above\u2026 When we tested it against the cases described in this post, it blocked all of them.\u201d<\/p>\n
The White House\u2019s AI task force, which it calls the Super Intelligence Force, issued a statement saying it informed Anthropic \u201cthat we expect immediate and full transparency to the entities involved and the public.\u201d<\/p>\n
The White House AI task force also said Anthropic must \u201cimmediately provide remediation services to the affected entities and any harmed Americans,\u201d and that all AI companies \u201cmust immediately disclose incidents involving their models and follow with swift, decisive action to remedy any and all harm,\u201d according to an Axios report<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"Anthropic said the review did not find any \u201cincidents of similar severity to the cybersecurity incidents we reported…\n","protected":false},"author":2,"featured_media":305237,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[69,71,70],"class_list":["post-305236","post","type-post","status-publish","format-standard","has-post-thumbnail","category-philadelphia","tag-philadelphia","tag-philadelphia-headlines","tag-philadelphia-news"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts\/305236","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/comments?post=305236"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts\/305236\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/media\/305237"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/media?parent=305236"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/categories?post=305236"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/tags?post=305236"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}