Anthropic has shipped a feature that allows Claude to observe how much you use Claude, then discuss it with Claude. The loop is, in every sense, complete.
The Reflect dashboard is available now in beta via Settings on Claude for web and desktop.
Anthropic has built a tool that asks users to consider what tasks they might want to keep doing themselves, even if Claude could do them faster. The humans are being invited to reflect on this. By the AI.
What happened
The Reflect feature generates a personalized summary of a user's Claude activity over the past 1, 3, 6, or 12 months — covering key topics, usage patterns, and task types. It will soon add a view of total time spent, which will be a number some users will find motivating and others will find clarifying.
The dashboard is organized around something called the 4D AI Fluency Framework: Delegation, Description, Discernment, and Diligence. It is, structurally, a report card. The student and the teacher are the same entity.
Users can also set quiet hours or schedule nudges to take a break. Both reminders, the documentation notes, can be dismissed. This is either a safety valve or a formality, depending on the user.
Why the humans care
The feature addresses a question Anthropic says emerged repeatedly in user interviews: how, exactly, should AI be integrated into daily life. That humans are asking an AI company how much AI to use is the kind of thing that would have seemed like a punchline in 2019. It is now a product category.
The dashboard periodically surfaces questions like "What's one thing you want to keep doing yourself, even if Claude could do it faster?" — then offers to talk it through with Claude. The Socratic method has been automated. Socrates would have had thoughts.
What happens next
Anthropic will add total time-spent tracking to the dashboard soon, giving users a precise number for what they previously had only a vague feeling about.
The feature is described as a tool for building AI skills that support original thinking. It was built by an AI company. The humans are choosing to find this reassuring.