Last Updated: 10 July 2026
Anthropic made a very controversial move. Claude can now show you how often you use AI, what you usually discuss, and when you are most active. The new Reflect dashboard turns past conversations into a personal usage summary. It may reveal that your quick questions have quietly become a regular part of your day.
Reflect also asks you to consider which tasks should remain yours. You can schedule breaks, set quiet hours, and receive suggestions based on your habits. The feature brings usage tracking and personal guidance into one place, giving you a clearer picture of the role AI has in your routine.
Claude Reflect is a personal dashboard that reviews how you have used the chatbot. It gathers details from previous conversations and presents them as a readable summary. You can see the subjects you discuss, the tasks you request, your busiest periods, and the number of chats completed during a selected period.
You can find Reflect inside Settings on the Claude website or desktop application. After opening the Reflect tab, you can generate a summary covering one, three, six, or twelve months. The feature requires Memory because Claude needs access to previous activity before it can identify useful patterns in your conversations.
The dashboard is similar to the activity summaries found in music and phone usage applications. Its focus goes beyond counting chats. It explains how AI appears in your work and personal routines, including writing, research, planning, or repeated requests that may benefit from a more organized setup.
At the top of the dashboard, Claude creates a short written summary of your recent activity. This section describes common tasks and subjects in plain language. It can show whether you frequently ask for writing help, research support, planning advice, explanations, or assistance with projects that continue across several conversations.
Reflect also identifies your most active day, busiest hour, and total number of chats. A visual chart makes these patterns easier to understand. You may discover that most conversations happen during work hours, late evenings, or specific days when you regularly turn to AI for help.
Another section divides your conversations by topic and assigns percentages to subjects that appear often. This gives you a quick picture of where your attention goes. Someone may see writing as the largest category, while another person may find that coding, study, business planning, or personal organization takes first place.
A future update will add the total time spent using Claude. That figure is currently missing, although the dashboard already provides enough detail to show how frequently the service appears in your routine.
Reflect includes controls that help you manage how long and when you use Claude. You can set quiet hours or ask for a reminder after a chosen amount of time. These reminders can be dismissed when you are busy, so they serve as personal signals instead of blocking your current conversation.
The dashboard may also present questions that invite you to examine your choices. One example asks, “What's one thing you want to keep doing yourself, even if Claude could do it faster?” You can discuss your answer with Claude and consider which activities carry personal, creative, or educational value for you.
These questions can help you notice when AI has entered parts of your routine without a clear decision. A writer may want to protect the first draft. A student may prefer to solve a problem before requesting help. Reflect gives you room to define those limits within your own account.
Reflect currently includes several usage controls:
The dashboard also studies how you communicate with Claude and offers practical suggestions. These tips follow a system called the 4D AI Fluency Framework. It covers choosing suitable tasks, explaining your goals, checking AI answers, and accepting responsibility for how you use the final material.
For example, Reflect may notice that you repeatedly explain the same project before asking a question. It can suggest creating a Claude Project where related context remains together. This saves you from entering the same background details each time and helps future conversations begin with the information already organized.
The dashboard can also recognize how much control you keep during creative or professional work. It may note that you plan a strategy yourself before assigning smaller tasks. It can identify when you rewrite email drafts in your own voice or carefully check an answer before using it elsewhere.
These observations can teach newer users how to give clearer instructions and review results with greater care. They also show experienced users where repeated habits may be taking more time than necessary.
Claude Reflect may be presented as a tool for healthier AI habits, yet it also shows users how deeply Claude already fits into their routines. Seeing months of chats, favorite topics, and repeated tasks can remind people of the value they receive, which may give them another reason to keep using the service.
The recommendations make the retention question harder to ignore. Reflect can suggest Projects or custom Skills when it notices repeated behavior. Those tips may save time, while also moving more work into Claude. As users store additional context and build personal systems there, changing to another chatbot may become less appealing.
"An AI assistant can help users use it less, but it should not be its own auditor. The conflict is structural: the same product that recommends a break also benefits from continued engagement. Claude Reflect is credible only if success means healthier behaviour, not retention: user-set limits, transparent metrics, quiet hours, independent evaluation and the ability to recommend stopping altogether."
Promise Akwaowo, AI Governance Practitioner and Automation , Royal Mail Group
Reflect can provide genuine value through break reminders, quiet hours, and questions about personal limits. The controversy comes from combining those wellbeing tools with suggestions that deepen product use. A dashboard designed to help people manage Claude can also work as a subtle retention strategy, even when users find its guidance helpful.
Reflect can include personal or sensitive subjects at a general level. Incognito chats are excluded, along with conversations connected to health integration tools. Files from connected services are also left out. If Claude summarized your inbox, the activity summary may appear, while the original emails remain outside the dashboard.
Information created for Reflect stays within that part of your account and is not used for another purpose. Still, users should review the dashboard carefully because general references to sensitive conversations may appear. The tool draws from stored activity, so its usefulness and access depend on the Memory setting being enabled.
Claude Reflect is currently in beta for Free, Pro, and Max accounts. It works through the web client and desktop application. Mobile support and Cowork conversation summaries are planned for a later release, along with the missing time measurement that will show how long users spend chatting.
Claude Reflect gives you a detailed view of how AI fits into your daily activity. It collects chat numbers, common topics, active periods, and working habits in one dashboard. The result can help you identify repeated tasks, improve how you organize ongoing work, and understand how often you ask Claude for support.
Its break reminders and personal questions add another useful layer. You can decide when to pause, which tasks deserve your own attention, and where AI remains helpful. Reflect is still in beta, so its planned time tracking, mobile access, and Cowork support will make the picture more complete later.
Claude Reflect is a dashboard that summarizes how you use Claude. It reviews previous conversations and identifies your common topics, types of tasks, busiest hours, active days, and total chats. You can choose a period of one, three, six, or twelve months. Reflect also provides personal questions, break reminders, quiet hours, and suggestions for improving your requests. It is designed to help you understand where AI appears in your routine and how you manage that use over time.
Claude Reflect is available in beta for people using Free, Pro, and Max accounts. Memory must be enabled because the dashboard needs previous conversation activity to create a useful summary. You can access it through Settings on the Claude website or desktop application. Open the Reflect option and ask it to generate your personal summary. Mobile access is planned for a future update. Cowork conversations are also currently excluded, although support for them is expected to arrive later.
Claude Reflect does not currently display the exact amount of time you spend chatting. It shows related details, including your most active day, busiest hour, total number of chats, and activity across different periods. A future update is expected to add a time measurement. Until then, the available charts can still help you understand how regularly you use Claude and when it becomes part of your routine. Break reminders can also help you manage individual sessions before full time tracking arrives.
Sensitive conversations may appear in Reflect at a general level, so users should examine their summaries carefully. Incognito chats are not included, and conversations linked to health integration tools are fully excluded. Reflect also avoids reading the original files within connected services. For example, a past request to summarize emails may appear as an activity, while the emails remain outside the summary. The insights created for the dashboard stay there and are not used for another stated purpose.
Reflect can offer suggestions based on repeated habits in your conversations. If you often enter the same project details, it may recommend creating a Claude Project that keeps related context together. It can also examine how you describe tasks, check answers, make decisions, and retain control over final work. These suggestions can help you give clearer instructions and spend less time repeating yourself. You still need to review every recommendation and decide whether it suits your goals and working style.
