How Claude's Memory Works: Personalization Across Conversations

What Claude's memory feature does, how to manage it, and the important distinction between memory for preferences and memory for sensitive information.

Projects give Claude context for a specific body of work. Memory gives Claude context about you across all conversations. As you use Claude over time, it can retain observations about your role, your communication preferences, the type of work you do, the terminology you use, and the formats you prefer. Those observations then carry forward into future conversations, making Claude progressively more useful the more you work with it.

  • What Claude's memory feature actually stores and how it builds over time
  • Where to find and manage your memory settings
  • How to give Claude explicit memory instructions
  • The difference between memory and projects, and how they work together
  • What memory is designed for and what it should never be used for

The following sections explain how memory works in practice, how to control it, and what the important limits are.

This lesson is a preview from our AI for the Workplace with Claude Course Online. Enroll in a course for detailed lessons, live instructor support, and project-based training.

Memory allows Claude to carry information about you from one conversation to the next. Rather than starting every session as a blank slate, Claude can remember things like your job title, your industry, the type of outputs you typically need, formatting preferences you have expressed, and terminology specific to your field. This means that over time, you spend less effort re-establishing context at the start of each conversation, because Claude already has a working understanding of who you are and how you like to work.

Managing Your Memory Settings

You have full control over Claude's memory. To access your settings, click on your profile icon in the lower portion of the interface and select Settings. From there, navigate to Capabilities and scroll down to the Memory section.

Within that section, you will find two toggles. The first controls whether Claude searches relevant details from past chats to inform its responses. The second controls whether Claude remembers relevant context from your chats and projects going forward. Both can be turned on or off at any time based on your preferences and needs.

If you want to view or edit what Claude has remembered about you, you can click into the Memory section and use the edit tool to review, add, or remove specific memory items. This level of control means you are never locked into information that is outdated or no longer accurate.

Giving Claude Explicit Memory Instructions

Beyond what Claude observes on its own, you can explicitly tell it what to remember. For example, you might instruct Claude to always use active voice in its responses, or to keep outputs under 300 words unless you specify otherwise. Instructions like these become part of Claude's persistent understanding of how you want it to behave, and they apply across future conversations without needing to be repeated.

Memory Versus Projects

Memory and projects serve related but distinct purposes. Projects give Claude specific context for a particular body of work: the files, instructions, and goals associated with one defined workspace. Memory gives Claude general context about you as a person and a professional: your role, your preferences, and how you like to communicate. The two work alongside each other. A project shapes how Claude behaves within that workspace. Memory shapes how Claude engages with you across all of your interactions.

What Memory Is For, and What It Is Not For

Memory is designed for professional context and preferences. It is appropriate for things like your job title, the industry you work in, your preferred output format, and the communication style you expect. It is not appropriate for sensitive personal information. Do not instruct Claude to remember passwords, social security numbers, confidential data, or any information that should not be stored in a third-party system. Memory is a productivity tool for professional context, not a secure vault for sensitive credentials or private data.

  • Memory allows Claude to retain context about your role, preferences, and working style across all conversations, not just within a single project.
  • Memory settings are found under Settings, then Capabilities, and can be toggled on or off at any time.
  • You can view, edit, and add to what Claude has remembered about you at any time.
  • Explicit memory instructions, such as formatting rules or style preferences, can be added directly and will persist across future conversations.
  • Memory provides general professional context. Projects provide specific workspace context. The two work together and serve different functions.
  • Memory should never be used to store sensitive personal information, passwords, or confidential data.
photo of Brian Simms

Brian Simms

Brian Simms teaches for Graduate School USA in the area of Artificial Intelligence, helping federal agencies build the knowledge and skills needed to adopt AI responsibly and effectively. An AI educator and author, he focuses on practical, mission-driven applications of AI for government leaders, program managers, and technical professionals.

More articles by Brian Simms

How to Learn AI

Build practical, career-focused skills in AI through hands-on training designed for beginners and professionals alike. Learn fundamental tools and workflows that prepare you for real-world projects or industry certification.