--- title: Meta Muse exposes users’ personal and business secrets... description: Meta's Muse AI agent exposes sensitive shared conversations through predictable, discoverable URLs. Security firm Claroc enumerated 6,800 shared conversation... canonical: https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Meta Muse exposes users’ personal and business secrets when they share conversations | daily.dev og:description: Meta's Muse AI agent exposes sensitive shared conversations through predictable, discoverable URLs. Security firm Claroc enumerated 6,800 shared conversation... og:url: https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b og:image: https://api.daily.dev/og/posts/S2brD3D4b.png og:image:alt: Meta Muse exposes users’ personal and business secrets when they share conversations og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Meta Muse exposes users’ personal and business secrets when they share conversations **[Ars Technica](https://daily.dev/sources/arstechnica)** · 3 min read · 0 upvotes · 0 comments ## Summary Meta's Muse AI agent exposes sensitive shared conversations through predictable, discoverable URLs. Security firm Claroc enumerated 6,800 shared conversation links using passive DNS searches against the domain cf.metaaiusercontent.com. Because Muse adds the conversation name to the URL along with a long string of text, researchers found loan applications, revenue reviews, bank fraud alerts, and a company-wide vacation calendar after opening a few hundred links. Claroc said gathering this data took only one query. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How did researchers find thousands of private Meta Muse shared conversations? Security firm Claroc used passive DNS searches against the domain cf.metaaiusercontent.com to enumerate 6,800 shared Muse conversation links. Muse embeds the conversation's title directly in the URL alongside a short random string, so titles like 'loan-application-draft' or 'q3-revenue-review' became searchable. Once a recipient clicked a share link, the DNS resolution was recorded and discoverable with a single query. _Developers designing shareable links track exposure risks like this one on daily.dev._ ### What sensitive data was exposed through leaked Meta Muse share links? Exposed conversations included a candidate's notes for a confidential executive-search call naming an AI lab's internal contact, a marketing company's internal revenue review marked for its CEO, a small-business loan application with the requested amount, a bank fraud alert with a partial account number, a real-estate investor's property returns and mortgage figures, and a full-year staff vacation calendar. _Anyone evaluating AI assistant privacy risks follows incidents like this on daily.dev._ ## Similar posts on daily.dev - [Meta’s Muse reportedly passes 3 million weekly users and 5m downloads](https://daily.dev/posts/meta-s-muse-reportedly-passes-3-million-weekly-users-and-5m-downloads-me2t3hekg) · The Next Web · 2 upvotes · 0 comments - [Meta debuts its Muse AI agent. Will consumers trust it?](https://daily.dev/posts/meta-debuts-its-muse-ai-agent-will-consumers-trust-it--nuqtze2hd) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Meta Muse exposes users’ personal and business secrets when they share conversations","url":"https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b"},"datePublished":"2026-10-20T12:50:57.924Z","dateModified":"2026-10-20T12:51:26.981Z","description":"Meta's Muse AI agent exposes sensitive shared conversations through predictable, discoverable URLs. Security firm Claroc enumerated 6,800 shared conversation...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e16bf0ff4f81f5e088f6b54661adb321?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e16bf0ff4f81f5e088f6b54661adb321?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Ars Technica","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Ars Technica","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/80883e0e48a34b5ebcf93777016cb3fe","url":"https://daily.dev/sources/arstechnica"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents","timeRequired":"PT3M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Ars Technica","item":"https://daily.dev/sources/arstechnica"},{"@type":"ListItem","position":3,"name":"Meta Muse exposes users’ personal and business secrets when they share conversations"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations-s2brd3d4b#faq","mainEntity":[{"@type":"Question","name":"How did researchers find thousands of private Meta Muse shared conversations?","acceptedAnswer":{"@type":"Answer","text":"Security firm Claroc used passive DNS searches against the domain cf.metaaiusercontent.com to enumerate 6,800 shared Muse conversation links. Muse embeds the conversation's title directly in the URL alongside a short random string, so titles like 'loan-application-draft' or 'q3-revenue-review' became searchable. Once a recipient clicked a share link, the DNS resolution was recorded and discoverable with a single query. Developers designing shareable links track exposure risks like this one on daily.dev."}},{"@type":"Question","name":"What sensitive data was exposed through leaked Meta Muse share links?","acceptedAnswer":{"@type":"Answer","text":"Exposed conversations included a candidate's notes for a confidential executive-search call naming an AI lab's internal contact, a marketing company's internal revenue review marked for its CEO, a small-business loan application with the requested amount, a bank fraud alert with a partial account number, a real-estate investor's property returns and mortgage figures, and a full-year staff vacation calendar. Anyone evaluating AI assistant privacy risks follows incidents like this on daily.dev."}}]} ```