Anthropic has introduced a new feature for its Claude chatbot that allows it to search a user’s chat history for information. According to the company, this function helps users seamlessly continue discussions by retrieving context from past interactions. The feature is currently rolling out to subscribers of the paid Max, Team, and Enterprise plans.
The search is activated only when a user explicitly prompts Claude to find something; it does not automatically scan chats for context. Users can ask natural language questions like, “What did we discuss about [topic]?” Claude can search across all chats (outside of designated projects) or be directed to search within a specific project’s conversations.
User control and privacy
Anthropic states the feature is enabled by default. However, users can disable it in their account settings under “Preferences”. To prevent a specific chat from being included in future searches, the user must delete that conversation entirely. This approach of user-initiated retrieval differs from other AI models that may automatically build “memories” of user details to personalize future responses.