Monday, October 12, 2026

AI & Models

Anthropic merges Claude's chat and Cowork memory into one system

Anthropic is merging Claude's chat and Cowork memory systems, so the assistant retains what it learns in one and lets users read, edit, or delete what it remembers.

A hand holding a smartphone displaying the Claude AI application splash screen with the Anthropic logo.
Photo: Anthropic

A change to Claude’s memory system will eliminate one of the most annoying things about using agents — the constant need to rebrief the AI on things it already knows. Anthropic announced Tuesday that it’s merging the memory system used by chat and Claude Cowork, meaning Claude will always remember what it learned in one area, even when a user is engaging with it in another. Anthropic is also exposing what Claude has retained in its memory, letting users read, edit, or delete information on any topic.

The update is meant to let users move more seamlessly between the research phase of a project and actually taking action. Previously, someone might develop ideas for a project over time while chatting with Claude, but when it came time to enact the changes, they’d have to start over by telling Cowork all the same details — an experience Anthropic frames as frustrating. The merge is meant to make Claude feel like one continuous assistant rather than two separate products.

Anthropic offers an example of someone drafting an update for their manager or an agenda for a conference, where Cowork would already understand references from past conversations — for instance, the headcount, city, and speakers involved — without the user re-entering that information. Claude will now also add topics to its memory as a conversation happens, rather than only summarizing a conversation once it ends, making memory updates move more quickly between chat and Cowork even while a chat is still ongoing.

Anthropic says that, by default, Claude won’t store personal or sensitive information such as health data, race, ethnicity, religious beliefs, politics, or gender identity, though users can opt in by toggling “include sensitive topics in memory,” and the app will notify users whenever a sensitive topic is saved. The company says Claude won’t ever save certain data, including government-issued ID and Social Security numbers, criminal history, or immigration status, or anything else that would violate its acceptable use policy.

The memory feature is enabled by default on Free, Pro, and Max plans across web, desktop, and mobile; iOS and Android users will need to update to the latest app version.

Why it matters

The merge treats Claude’s memory as a single, user-controllable layer rather than per-product silos, which could make the assistant more useful for people who move fluidly between planning work in chat and executing it in Cowork.