--- title: Claude's memory now works the same way across chat and... description: Anthropic unified Claude's memory system so it behaves the same in regular chat and in Cowork sessions, closing a gap where Cowork previously lacked persistent... canonical: https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Claude's memory now works the same way across chat and Cowork | daily.dev og:description: Anthropic unified Claude's memory system so it behaves the same in regular chat and in Cowork sessions, closing a gap where Cowork previously lacked persistent... og:url: https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja og:image: https://api.daily.dev/og/posts/QGPZM53Ja.png og:image:alt: Claude's memory now works the same way across chat and Cowork 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. # Claude's memory now works the same way across chat and Cowork **[Collections](https://daily.dev/sources/collections)** · 2 min read · 2 upvotes · 0 comments ## Summary Anthropic unified Claude's memory system so it behaves the same in regular chat and in Cowork sessions, closing a gap where Cowork previously lacked persistent memory. Memory now updates live during long sessions rather than only after they end, and memories are stored as editable files under Topics in Memory settings. Privacy defaults are conservative: sensitive categories like health, religion, race, ethnicity, and gender identity require opt-in, while SSNs, government IDs, criminal history, and immigration status are never stored. Memory is on by default for Free, Pro, and Max plans but off by default for Enterprise and Teams, with a 30-day retention rule and potential future self-hosted storage for enterprises. ## Content Anthropic has merged the memory systems used by Claude's chat and Claude Cowork, so context built up in one carries automatically into the other. Previously, switching from a chat session to an agentic Cowork task meant re-briefing Claude on project details, preferences, or team definitions. Now those carry over without any extra prompting, and updates flow in both directions. A few other changes came with this: - Memory is now built incrementally during conversations rather than only summarized at the end - Users can view, edit, or delete individual stored topics under Settings > Memory, which now has its own dedicated tab - You can tell Claude what to remember directly from the memory browser UI - Stored memories are organized into separate topic categories ## What gets stored and what doesn't Sensitive topics — health, religion, race, ethnicity, politics, gender identity — are excluded from memory by default. There's a new toggle in Settings to opt into storing those if you want. Some data is never stored regardless of settings: Social Security numbers, government IDs, criminal history, and immigration status. When a sensitive topic does get saved, Claude notifies you, and those memories only apply going forward. ## Who it's on for Memory is enabled by default for Free, Pro, and Max plans across web, desktop, and mobile (iOS and Android users need to update to the latest app version). Team and Enterprise accounts have it off by default, controlled by admins. Claude Code is not part of this update. ## The security angle worth noting Some observers have pointed out that unifying memory with Cowork raises the stakes compared to chat-only memory. Cowork can execute tasks autonomously, and it had a sandbox-escape vulnerability in the past. A persistent memory layer that feeds into an agent capable of taking real-world actions is a wider surface for things to go wrong. This update also coincides with Cowork expanding to Chrome and mobile, and with competitive pressure from Cursor's similar agentic product. One limitation that hasn't changed: memory files aren't portable. If you want to switch to a different AI provider, your stored context stays with Anthropic. ## Questions this post answers ### Does Claude's Cowork mode now have persistent memory like regular chat? Yes, Anthropic merged the memory systems so Cowork sessions carry the same persistent memory as regular chat conversations. Previously Cowork lacked this feature, so the assistant could forget a manager's preferences or a project's headcount between sessions. Memory now also updates live during long Cowork sessions instead of only after a conversation ends, and corrections made mid-session persist into future conversations. _Teams weighing AI coworker tools for continuity across sessions can track updates like this on daily.dev._ ### What personal data will Claude refuse to save in memory even if I opt in? Social security numbers, government IDs, criminal history, and immigration status are never stored in Claude's memory, regardless of opt-in settings. Other sensitive categories like health, religion, race, ethnicity, and gender identity are only saved if a user enables the 'include sensitive topics in memory' setting, and a notice appears whenever a sensitive topic gets saved. _Developers evaluating AI tool data handling policies can follow privacy-related changes like this on daily.dev._ ### Is Claude's memory feature turned on by default for Enterprise and Teams accounts? No, memory is off by default for Enterprise and Teams plans, while it's on by default for Free, Pro, and Max users. Anthropic applies a 30-day retention rule for stored memories, and there are indications enterprises may eventually get an option to keep memory data on their own infrastructure instead of leaving it with Anthropic. _Teams choosing AI tools for sensitive organizational data can track policy shifts like this on daily.dev._ --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#data-privacy](https://daily.dev/tags/data-privacy), [#claude](https://daily.dev/tags/claude), [#anthropic](https://daily.dev/tags/anthropic) [View this post on daily.dev](https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja) ```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":"Claude's memory now works the same way across chat and Cowork","url":"https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja"},"datePublished":"2026-08-25T18:15:34.418Z","dateModified":"2026-08-26T21:09:24.298Z","description":"Anthropic unified Claude's memory system so it behaves the same in regular chat and in Cowork sessions, closing a gap where Cowork previously lacked persistent...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/9bc30dd52b56cb7466a4862f4deccd13?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/9bc30dd52b56cb7466a4862f4deccd13?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Collections","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":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,data-privacy,claude,anthropic","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Claude's memory now works the same way across chat and Cowork"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/claude-s-memory-now-works-the-same-way-across-chat-and-cowork-qgpzm53ja#faq","mainEntity":[{"@type":"Question","name":"Does Claude's Cowork mode now have persistent memory like regular chat?","acceptedAnswer":{"@type":"Answer","text":"Yes, Anthropic merged the memory systems so Cowork sessions carry the same persistent memory as regular chat conversations. Previously Cowork lacked this feature, so the assistant could forget a manager's preferences or a project's headcount between sessions. Memory now also updates live during long Cowork sessions instead of only after a conversation ends, and corrections made mid-session persist into future conversations. Teams weighing AI coworker tools for continuity across sessions can track updates like this on daily.dev."}},{"@type":"Question","name":"What personal data will Claude refuse to save in memory even if I opt in?","acceptedAnswer":{"@type":"Answer","text":"Social security numbers, government IDs, criminal history, and immigration status are never stored in Claude's memory, regardless of opt-in settings. Other sensitive categories like health, religion, race, ethnicity, and gender identity are only saved if a user enables the 'include sensitive topics in memory' setting, and a notice appears whenever a sensitive topic gets saved. Developers evaluating AI tool data handling policies can follow privacy-related changes like this on daily.dev."}},{"@type":"Question","name":"Is Claude's memory feature turned on by default for Enterprise and Teams accounts?","acceptedAnswer":{"@type":"Answer","text":"No, memory is off by default for Enterprise and Teams plans, while it's on by default for Free, Pro, and Max users. Anthropic applies a 30-day retention rule for stored memories, and there are indications enterprises may eventually get an option to keep memory data on their own infrastructure instead of leaving it with Anthropic. Teams choosing AI tools for sensitive organizational data can track policy shifts like this on daily.dev."}}]} ```