Anthropic Unifies Memory Across Claude and Cowork

Anthropic has enhanced memory capabilities across its AI tools, Claude and Cowork. When switching between the two, users will retain the same memory from past chats, eliminating the need to reiterate context and saving valuable time. This upgrade allows users to focus on their projects without interruption.

In addition to this seamless transition, Anthropic has introduced a range of memory-related features. Users can now specify what information they want to store and manage their chat history more efficiently. Claude updates its memory in real-time as conversations progress, enabling it to recall specific details without requiring explicit instructions.

Managing memories is now simpler than ever. Users can view all topics Claude has saved from previous chats and make edits or deletions within the memory settings. Additionally, Claude can now store sensitive topics that were previously excluded. Users have the option to enable this feature, which can be beneficial for remembering personal preferences, such as dietary restrictions.

While the default setting allows memory across free, Pro, and Max plans on web, desktop, and mobile platforms, certain sensitive information, including government ID numbers, criminal history, financial account numbers, and immigration status, will not be automatically stored. These topics remain off by default and must be manually enabled.

Sensitive topics have a separate setting that, when enabled, ensures Claude remembers important details, like a user's gluten allergy when generating meal prep recipes. However, even with this setting enabled, some highly sensitive information remains protected and will not be stored in memory. Claude will notify users when it is unable to save sensitive data.

For mobile users, Anthropic has updated the Claude app for both iOS and Android to incorporate these new memory-enhancing features.