Meet Claude Wrapped

The widespread success of Spotify Wrapped has inspired a wave of year-in-review functionalities across various applications, from YouTube to Uber. This trend of digital introspection has now extended into the realm of artificial intelligence. Anthropic recently unveiled a new "reflect" feature for its Claude chatbot, which the company terms a "reflection dashboard." This innovative tool offers users a comprehensive analysis of their interaction data with Claude over periods of one month, three months, six months, or an entire year.

Anthropic positions this reflection dashboard as a mechanism for users to "see your patterns and shape them," as detailed in a recent blog post. The feature commences by presenting a summary of an individual's primary discussion topics with Claude, the nature of tasks they delegate to the AI, and their overall usage patterns, including peak activity times. Furthermore, users will gain the ability to designate "quiet hours" or configure reminders for breaks after specific durations of use. Anthropic also indicated that a metric detailing total time spent utilizing Claude, likely an aggregated hours-of-usage tally, is slated for future implementation.

The company elaborates on the feature's intent, stating, "It lets you easily track and visualize how you use Claude, and decide whether that time aligns with your goals… Your reflection also invites you to step back and examine the role Claude plays in your life." The dashboard will periodically pose thought-provoking questions, such as, "What’s one thing you want to keep doing yourself, even if Claude could do it faster?" Intriguingly, Anthropic noted that once a user provides an answer to such a question, the dashboard will then offer them "the chance to talk it through with Claude."

Anthropic has significantly invested in promoting itself as the developer of an "AI collaborator" designed to foster deeper human thinking, evidenced by extensive advertising campaigns spanning ads, billboards, and even novelty "thinking" caps. The "reflect" feature is presented with the same philosophy, with the company's blog post encouraging users to leverage it to "build AI skills that support your original thinking." This functionality will highlight specific interaction examples, such as observing that a user frequently refines email drafts in their personal voice or only delegates tasks after independently establishing the core strategy.

Regarding data privacy, Anthropic clarified that the reflection dashboard will not access files from connected tools or platforms, meaning it won't pull actual emails from a linked email account. However, the fact that a user might have employed Claude to summarize their email inbox could be noted in the reflection. Similarly, the feature will not incorporate chats conducted in incognito mode or make any reference to conversations linked to a "health integration tool." While Anthropic remains somewhat vague in defining "sensitive topics," the company stated that "sensitive conversations can still appear as part of your reflection, but only at a high level."

Currently available in beta, this feature is accessible to both free users of Claude and subscribers of its Pro and Max tiers. It requires users to have the memory function activated for their chats and can be accessed through the Settings menu on either the web interface or the Claude desktop application. Anthropic revealed that the concept for this feature arose from direct interviews with Claude users, and it is anticipated to be rolled out to Claude Cowork in the near future.