Claude memory in Cowork

Claude memory is a feature of Anthropic's Claude apps that stores facts and context about a user so later conversations start from what Claude already knows. Anthropic's announcement states that the memory used in chat is now the same one used in Claude Cowork, so context from chats is available when Cowork runs a task in the cloud, and what comes up in Cowork carries back to chat.\n\nClaude now adds topics to memory while the user chats, rather than summarizing conversations after they end. Everything remembered is shown as a list of short files under Topics in Settings > Memory, where each can be read, edited or deleted. Memory can also be paused or reset at any time.\n\nSensitive topics (for example health, race, ethnicity, religious beliefs, politics, gender identity) are not stored by default. Users can enable "include sensitive topics in memory"; Claude then shows a notice each time it saves such an item, and nothing from before the setting was enabled is saved retroactively. Some content is never stored even with the setting on, including identification numbers such as SSNs, criminal history, immigration status, and anything violating the Acceptable Use Policy.\n\nMemory is on by default on Free, Pro, Max and Ultra plans across web, desktop and mobile. For Team and Enterprise, admins control availability; per an update, memory is on by default for members of most organizations created on or after October 31, 2026, and owners can turn it off in Organization settings.