Anthropic's Claude AI chatbot takes a leap forward by introducing a dynamic memory system designed specifically for workplace collaboration. This transformative update promises to eliminate repetitive context-setting and streamline complex projects, redefining how teams and enterprises interact with AI.

Memory that Remembers You and Your Team

Starting today, Claude users on the Team and Enterprise plans can benefit from a new memory feature where the AI “remembers” details from previous conversations — including project specifics, client preferences, and workflows. This means you no longer have to re-explain context every time you start a new chat, allowing for smoother multi-session collaboration.

Unlike generic chatbots that treat every conversation as a fresh start, Claude’s memory is purpose-built for the workplace. It learns and retains relevant professional context, helping sales teams recall client needs across negotiations, product teams maintain specs over multiple sprints, and executives keep track of ongoing initiatives — all without repeated reminders.

Project-Centric Memory: Keeping Conversations Organized and Secure

Recognizing the complexity of juggling multiple projects, Claude automatically creates separate memories for each project. This clean separation ensures sensitive client details, product development plans, and internal operations stay compartmentalized, providing an added layer of confidentiality and clarity.

This approach safeguards against accidental mixing of unrelated information, making it easier for teams to manage parallel tasks without confusion. Enterprise administrators retain full control with the power to disable memory at any time, ensuring organizations maintain governance over their data.

Granular Control and Transparency

Privacy and user autonomy are paramount. Claude’s memory is fully optional and highly customizable:

• View and Edit: Users can review everything Claude has retained via a memory summary and update it on the fly by simply chatting with the AI.

• Focus & Ignore: By telling Claude what to remember or disregard, the AI adjusts accordingly, tailoring its recall to match user preferences.

• Incognito Mode: For sensitive discussions, Claude offers Incognito chats that do not save history or add to memory, perfect for confidential brainstorming or strategy sessions. This mode is available to all users, regardless of plan.

How to Get Started with Claude Memory

To try memory in Claude, just enable it in Settings. At setup, Claude can generate memory from past chats. You can then ask questions like “What were we working on last week?” to see what it remembers across chats and tools. If you want to import memory from another AI tool or export Claude’s memory for backup/migration, check the details here: Importing and Exporting Your Memory from Claude.

Announcement