Anthropic is trying to erase one of AI assistants’ most familiar irritations: having to explain the same project twice. Its Claude Cowork update gives the work-focused tool access to the memory already built in regular Claude chats.

The change, announced Tuesday, merges the previously separate chat and Cowork memory systems. That means a user who has discussed a conference, a team’s headcount or a draft project in Claude’s chat can move into Cowork without rebuilding the context from scratch. One account of the rollout called the old handoff “incredibly frustrating,” arguing that the update makes Claude feel “like one continuous assistant, rather than two separate products under one roof.”¹

The practical pitch is continuity: research and brainstorming can happen in chat, then flow into an action such as drafting a manager update or conference agenda in Cowork. Claude will also add topics to memory as conversations continue, rather than waiting to summarize them at the end, speeding the sharing of new context between the two experiences.¹

But the convenience comes with an explicit privacy boundary. Memory is enabled by default across Free, Pro and Max plans on web, desktop and mobile, though mobile users must update their apps. Anthropic says sensitive categories — including health, race, religion, politics and gender identity — are excluded by default. Users may opt in to storing them and will be notified when such material is saved; certain data, including government IDs, Social Security numbers, criminal history and immigration status, will not be stored.¹

Users can inspect, edit or delete stored information through Claude and Cowork settings. As a second report put it, the shared system is designed to make it easier to resume tasks “without re-explaining important information,” while leaving sensitive-memory controls in users’ hands.²