Anthropic has launched Claude Reflect, a new dashboard that helps users understand how they use the company’s AI assistant over time. Available in beta starting today, the feature combines usage analytics, productivity insights, and digital wellbeing tools to give users a clearer picture of how Claude fits into their daily workflow.
Rather than simply showing activity statistics, Claude Reflect is designed to answer broader questions about AI usage. Anthropic says many users want to better understand when AI is most useful, how often they rely on it, and which tasks are still better handled without AI. Reflect was built to help users evaluate those habits while offering suggestions to improve the way they collaborate with Claude.
Claude Reflect generates a personalized report based on your conversations from the past one, three, six, or twelve months. The dashboard summarizes the topics you discuss most often, the types of tasks you complete with Claude, and when you tend to use the assistant throughout the day. It also displays your most active day, peak usage hour, total chats, and a visual activity timeline. Anthropic says a future update will add the total amount of time you’ve spent using Claude.
The dashboard also includes a breakdown of the subjects you discuss most frequently, making it easier to see how your conversations are distributed across different topics. Whether you primarily use Claude for writing, coding, research, or brainstorming, Reflect provides a high-level overview of where your AI interactions are focused.
Beyond usage statistics, Claude Reflect periodically asks questions designed to encourage self-reflection. For example, it may ask users, “What’s one thing you want to keep doing yourself, even if Claude could do it faster?” The goal is to help people think more intentionally about where AI adds value and where human judgment remains important.
To support healthier AI habits, Claude Reflect also introduces new wellbeing controls. Users can set Quiet Hours or schedule reminders that encourage them to take a break after using Claude for a certain amount of time. These reminders reflect the user’s own preferences and can be dismissed whenever needed.
Another key part of Claude Reflect is its integration with Anthropic’s 4D AI Fluency Framework, which evaluates how people work with AI across four areas:
- Delegation: Deciding when and how to hand work over to AI.
- Description: Writing clear prompts that produce better results.
- Discernment: Evaluating AI responses and determining whether they are accurate and useful.
- Diligence: Taking responsibility for how AI-generated information is used.
Based on your conversations, Claude Reflect provides examples of your working style and practical recommendations. For example, if you repeatedly explain the same project context, Claude may recommend using Projects to organize ongoing work instead of starting from scratch each time. It can also suggest creating custom workflows or skills for repetitive tasks, helping users get more consistent results from Claude.
Anthropic says privacy was a major consideration while developing Claude Reflect. The feature does not analyze Incognito chats, and it does not access the original files connected through external tools. For example, if Claude summarizes your inbox, that summary may appear in Reflect, but the original emails are never accessed or displayed. Conversations connected to health integration tools are excluded entirely, and Anthropic says the information generated by Reflect remains within the dashboard and is not used for any other purpose.
The company also worked with researchers from the MIT Media Lab’s Advancing Humans with AI program, Boston Children’s Hospital’s Digital Wellness Lab, and the Family Online Safety Institute while developing the feature. Even when sensitive conversations appear in Reflect, they are summarized only at a high level rather than exposing detailed personal information.
Claude Reflect is currently available in beta for Free, Pro, and Max users who have Memory enabled. If Memory is turned off, users will not be able to generate a report. The feature is available through the Reflect tab in the Settings menu on Claude for the web and desktop app. Anthropic says support for Claude Cowork conversations, mobile access, and detailed time tracking will be added in future updates.
(via Anthropic)
