Anthropic just launched Claude's Reflect dashboard, a feature that visualizes your AI usage patterns. It sounds like a wellness tool. It functions like a loyalty program. The more you see how deeply you rely on Claude, the harder it becomes to imagine life without it. This is not incidental design. It is the product.
Quantified Self as Sales Funnel
The Reflect dashboard joins a long lineage of self-tracking tools that convert behavior into dependency. Fitbit made you addicted to your step count. Spotify Wrapped made you addicted to your listening identity. Now Anthropic is making you addicted to your own productivity narrative, with Claude as the hero of the story. A 2023 paper in Computers in Human Behavior by Rooksby et al. found that personal informatics tools systematically shift user self-concept toward the metrics being tracked. When the metric is AI dependence, the feedback loop becomes a revenue loop.
When the Product Becomes the Ideology
This connects directly to the staggering scale of pending AI IPOs, where Anthropic, OpenAI, and SpaceX are projected to generate more value than all U.S. VC-backed exits since 2000. At that scale, the product is not the chatbot. The product is the belief system around AI as infrastructure. Reflect is not a dashboard. It is a conversion event. Meanwhile, the arXiv paper on LLM moral reasoning by Fourie, Ray, and Manicom found that user identity framing shapes how AI systems assign moral weight, which means these tools are not neutral mirrors at all. They reflect a version of you that justifies continued use. Kyle Chayka's analysis of algorithmic homogenization applies here with uncomfortable precision: the feed does not show you who you are, it shows you who it needs you to be. Anthropic just built that into a wellness feature.