Anthropic Unifies Claude Memory Across Chat and Cowork with User-Controlled Privacy Settings
Key point
Memory is enabled by default for Free, Pro, and Max plans, while sensitive topics require explicit user activation.
Details
Starting August 25, 2026, Claude uses a unified memory system across Claude Cowork and chat interfaces. This integration ensures that context built in one environment is immediately available in the other, reducing the need for users to re-explain preferences, project statuses, or personal details.
Real-time Updates and User Control
Unlike previous systems that summarized conversations after they ended, Claude now adds topics to memory in real-time as users chat. Users can view, edit, or delete specific memory files under Topics in Memory settings, allowing for precise corrections that propagate across all future interactions. For example, correcting a company name in one file ensures accuracy in every subsequent conversation.
Privacy and Sensitive Topics
By default, Claude does not store sensitive subject matter such as health, race, ethnicity, religious beliefs, politics, or gender identity. Users can opt-in to "include sensitive topics in memory" to allow Claude to remember details like dietary restrictions for meal planning. Even with this setting enabled, Claude will never store sensitive identification numbers (SSN, government IDs), criminal history, or immigration status.
Availability
Memory is on by default for Free, Pro, and Max plans on web, desktop, and mobile. For Team and Enterprise users, admins control availability, and memory remains off for individual users until they manually enable it. iOS and Android users must update to the latest app version to access these features.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.