{"id":674206,"date":"2026-11-04T01:04:12","date_gmt":"2026-11-04T01:04:12","guid":{"rendered":"https:\/\/www.newsbeep.com\/ie\/674206\/"},"modified":"2026-11-04T01:04:12","modified_gmt":"2026-11-04T01:04:12","slug":"anthropic-is-working-on-a-system-to-prove-ai-models-identity","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ie\/674206\/","title":{"rendered":"Anthropic is working on a system to prove AI models&#8217; identity"},"content":{"rendered":"<p id=\"speakable-summary\" class=\"wp-block-paragraph\">Anthropic is working on an open source system to verify which AI model produced a given output, according to a blog post published Tuesday. The system, <a href=\"https:\/\/www.anthropic.com\/research\/provable-inference\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">called FlashProve<\/a>, is built to provide what researchers call \u201cprovable inference\u201d \u2014 a cryptographic proof, essentially a mathematical guarantee, that a given output came from a specific model. It\u2019s meant to let users confirm that the answer they received really came from the model they think they\u2019re using.<\/p>\n<p class=\"wp-block-paragraph\">Similar to <a href=\"https:\/\/en.wikipedia.org\/wiki\/DomainKeys_Identified_Mail\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">cryptographic email signatures<\/a> that let recipients confirm a message really came from its claimed sender, FlashProve would be robust against impostors attempting to disguise one model as another, even when the impostor is the lab that built and runs the model. In particular, the researchers note that the system could be used to prevent an AI company from secretly swapping in a cheaper model to cut costs, or a more expensive model to inflate its performance results.<\/p>\n<p class=\"wp-block-paragraph\">FlashProve could also be used to verify claims by third parties that a specific task was done by an advanced model. \u201cA developer might claim that their software library was scanned for vulnerabilities by a frontier cybersecurity model, or a company might claim that its business plan was given the go-ahead by an appropriate model,\u201d the researchers write. \u201cHow can a user verify these claims?\u201d<\/p>\n<p class=\"wp-block-paragraph\">To work, the system requires AI labs to publish a \u201cfingerprint\u201d of each model they release that can later be matched to specific outputs and verified cryptographically. Crucially, the fingerprints don\u2019t reveal a model\u2019s weights, the internal numbers that define how it behaves and which closed-model makers like Anthropic and OpenAI keep private. That means labs could publish the fingerprints without giving up their most valuable asset, enabling the system to work with commercial models, too. The system is also designed to be lightweight, so verification could be performed on a local device.<\/p>\n<p class=\"wp-block-paragraph\">As part of the announcement, researchers released <a href=\"https:\/\/assets.claude.com\/8e1d181a83015078581c92d896757041713f0160.pdf\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">a detailed technical specification<\/a> and working code for various elements of the system, which is publicly available <a href=\"https:\/\/github.com\/anthropics\/FlashProve\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">under an open source license on GitHub<\/a>. Research in the system is ongoing, and there are no plans to deploy it at Anthropic or any other lab.<\/p>\n<p class=\"wp-block-paragraph\">\u201cThis research post is meant to share intermediate progress, and not a completed research agenda,\u201d researchers explain in the announcement post. \u201cWe are sharing this explicitly with the goal that others look at what we have done, take it, and improve it.\u201d<\/p>\n<p>When you purchase through links in our articles, <a href=\"https:\/\/techcrunch.com\/techcrunch-affiliate-monetization-standards\/\" rel=\"nofollow noopener\" target=\"_blank\">we may earn a small commission<\/a>. This doesn\u2019t affect our editorial independence.<\/p>\n","protected":false},"excerpt":{"rendered":"Anthropic is working on an open source system to verify which AI model produced a given output, according&hellip;\n","protected":false},"author":2,"featured_media":674207,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[220,3673,218,219,11570,61,16374,60,80],"class_list":["post-674206","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-anthropic","tag-artificial-intelligence","tag-artificialintelligence","tag-cryptography","tag-ie","tag-inference","tag-ireland","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/674206","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=674206"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/674206\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/674207"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=674206"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=674206"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=674206"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}