Claude Cowork now remembers what you told Claude in chat
What's the story
Anthropic has announced a major update to its AI assistant, Claude. The company is integrating the memory system of Chat and Claude Cowork, making it possible for the AI to remember information across different conversations. This means you won't have to keep repeating things you've already told the app. The update is expected to enhance user experience by allowing seamless transitions between research and action phases.
Memory management
Users can edit or delete stored information
Along with the update, Anthropic is also giving users more control over what Claude remembers.
You can now view, edit, or delete any information stored in the AI's memory.
This feature is designed to give you more power over your interactions with Claude and make it a more personalized experience.
Memory enhancement
Memory will now expand during conversations
The updated version of Claude will now expand its memory during conversations instead of summarizing them at the end.
This change will make switching between chat and Cowork easier, as new memories are updated and shared quickly between both experiences, even if the chat is still ongoing.
The feature is enabled by default on Free, Pro, and Max plans across web, desktop, and mobile platforms.
Memory sensitivity
What's remembered and what's not
By default, Claude doesn't remember personal or sensitive information such as health data, race, ethnicity, religious beliefs, politics, gender identity.
However, you can choose to include these by enabling "include sensitive topics in memory."
The app will also alert you whenever a sensitive topic is added to its memory.
But some things like government ID and Social Security numbers won't be saved at all due to violation of acceptable use policy.