Anthropic has released, in beta, a new feature that lets you reflect on and rethink how you use its conversational AI, Claude[1]. From the settings screen, you can visualize your own usage and check whether the time you spend with AI aligns with your goals. It is a feature built to support the judgment of the person using it.
A "Reflect" dashboard available from Settings
The feature lives under "Reflect" in Claude's Settings on the web or the desktop app[1]. According to Anthropic, interviews with users surfaced a common interest: how, exactly, to fit AI into daily life. How often should you use it? Which tasks are suited to AI, and which are better left to a person? The tool was built to help answer questions like these[1].
The reflection dashboard starts with a summary of how you have been using Claude. It gathers your most frequent topics, your usage patterns, and the kinds of tasks you take on, and it lets you look back over the past 1, 3, 6, or 12 months[1]. It also shows when you use Claude most and what you were working on during that time, and a view of how much time you have spent using Claude will be added soon[1].
Prompts to question your habits, and reminders to take a break
The feature goes beyond simply displaying a usage log. So that you can step back and re-examine the role Claude plays in your life, it periodically raises questions such as "What's one thing you want to keep doing yourself, even if Claude could do it faster?" and lets you talk them through with Claude on the spot[1].
Within the dashboard, you can also set a nudge that suggests a break after a certain amount of use, or set "quiet hours" during which you step away from Claude. Both are meant to remind you of preferences you set yourself, and you can dismiss them if you don't need them[1]. Rather than imposing a blanket limit on heavy use, the design leans toward supporting the user's own decisions.
Sharpening how you work with AI through the "four Ds"
Your reflection can also be used to build the skills for working well with AI. Anthropic points to four perspectives (the 4Ds) as a guide to AI fluency, the ability to use AI effectively and safely[1]. They are Delegation (judging whether to hand a task to AI), Description (conveying your goals precisely), Discernment (correctly judging how useful an output is), and Diligence (taking responsibility for how you use AI)[1].
The report summarizes your tendencies across each of these four dimensions, offering concrete examples such as "you often rework email drafts in your own words" or "you delegate tasks only after settling your strategy." It also suggests practical improvements, like starting a "Project" instead of re-explaining your context every time[1].
Attention to privacy, and who can use it
Because this is a feature that compiles your usage, Anthropic also explains its approach to privacy. The reflection does not draw on the contents of incognito chats, nor does it reference the underlying files pulled in from connected tools[1]. For example, if you asked Claude to summarize your inbox, that summary may appear in your reflection, but the original emails themselves are not included. Conversations connected to a health-related integration tool are excluded entirely[1]. The information and insights gained from the reflection stay there and are not used for any other purpose[1].
On handling sensitive topics, Anthropic says it designed the feature with experts including the Advancing Humans with AI (AHA) program at the MIT Media Lab, the Digital Wellness Lab at Boston Children's Hospital, and the Family Online Safety Institute. Sensitive conversations may still appear in your reflection, but only at a high level[1].
The feature is available to Free, Pro, and Max plan users who have the memory feature turned on, starting as a beta[1]. Open Settings in Claude on the web or the desktop app and choose the option to reflect on your usage, and a report will be generated. If you can't create a report, it may be because memory is not enabled. A reflection covering conversations in Cowork is also set to arrive soon[1].
Summary
Claude's new reflection feature is meant to help users grasp how much and in what ways they use AI, and to rethink how they engage with it. Beyond visualizing usage trends, it includes break reminders, a prompt asking what you want to keep doing yourself, and a review of your habits through four perspectives. It is a feature that suggests a stance of ultimately leaving the decision of how to fit AI into life and work to the person using it.
