Anthropic Unifies Claude Memory Across Chat and Cowork

Anthropic has unified Claude memory across Chat and Cowork, allowing the same memory to be shared between both experiences. The update means users spend less time re-explaining projects, preferences, and ongoing work. Anthropic announced the change in a post on the official Claude blog.



Previously, context built in Chat did not automatically carry over to Cowork. Now, anything Claude remembers from conversations is available when Cowork performs cloud-based tasks, and information gathered by Cowork carries back into Chat.


The shared memory lets Claude draw on context built across months of conversations, including project status, recurring priorities, and how specific people prefer updates to be written. Cowork can draft an update while already knowing who your manager is and how that person prefers updates to be written. Likewise, a conference budget and logistics document can already include details such as the headcount, city, and speakers discussed earlier in Chat.



Anthropic is also changing how Claude updates memory. Instead of waiting until a conversation ends, Claude now adds topics to memory as conversations happen. Mention that a project deadline moved to September, and the next conversation already reflects the change without requiring a separate "remember this" prompt. Users can also pause or reset memory at any time.

Everything Claude remembers appears as a list of short files under Topics in Memory settings. Each one can be viewed, edited, or deleted individually, with changes applied across future conversations. Correct a company's old name in one topic, for example, and Claude uses the updated information going forward.

By default, Claude does not save sensitive topics such as health information, race, ethnicity, religious beliefs, politics, or gender identity. Users who want Claude to retain that information can enable the "include sensitive topics in memory" setting, and Claude displays a notice whenever it saves sensitive content. Anything discussed before the setting is enabled is not saved retroactively. Some information is never stored, including Social Security numbers, government identification numbers, criminal history, immigration status, or anything that violates Anthropic's Acceptable Use Policy.


The feature is enabled by default for Free, Pro, and Max users across the web, desktop, and mobile apps. iPhone and Android users need to update to the latest version of the mobile app. For Team and Enterprise plans, administrators control availability for their organizations, and memory remains off for individual users until they turn it on in Settings > Memory.