Detailed Analysis
A Reddit user's complaint about Claude's "Move to Project" feature highlights a gap between user expectations and actual functionality in Anthropic's chat organization system. The user bulk-selected several conversations from their chat history and used the "Move to project" option, anticipating that these chats would be relocated exclusively into the designated project folder—removed from the general Recents view and main chat history. Instead, the chats appeared simultaneously in three locations: the project, Recents, and the full chat history list. This behavior suggests that "Move to Project" functions more as a tagging or association mechanism than a true relocation feature, despite the terminology implying otherwise.
This discrepancy matters because it touches on a fundamental usability expectation in software design: verbs like "move" carry strong semantic implications. Users generally understand "move" to mean an item is relocated from one place to another, not duplicated or referenced in multiple places simultaneously. When a product's actual behavior contradicts this expectation, it creates confusion and erodes trust in the interface, especially for power users who rely on organizational systems to manage large volumes of content. The original poster's observation that there are no visible badges or indicators showing which chats have been associated with a project compounds the problem—without any visual cue, users have no way to distinguish "organized" chats from ordinary ones sitting in Recents, effectively negating the value of the categorization effort entirely.
This issue reflects a broader pattern seen across many AI chat platforms, including ChatGPT and other LLM interfaces, where features for organizing conversations into folders, projects, or workspaces often lag behind the core conversational capabilities in terms of polish and intuitive design. As Anthropic's Claude has evolved from a simple chatbot into a more comprehensive productivity tool—with Projects functioning similarly to workspaces that can hold context, files, and multiple related conversations—users increasingly expect file-system-like behaviors: moving, archiving, and nesting items with predictable outcomes. The friction described here suggests that while Anthropic has built the scaffolding for project-based organization, the underlying information architecture (whether a chat has one "home" or can have multiple parent associations) may not have been fully reconciled with user-facing labels and controls.
More broadly, this kind of feedback is emblematic of the growing pains AI companies face as their products mature from novelty tools into daily-use productivity software. As users generate increasingly large volumes of conversation history through sustained, repeated use, the demand for robust knowledge-management features—search, tagging, folders, archiving—grows in parallel. Anthropic, like its competitors, will likely need to invest more deliberately in these "boring" but essential UX layers, since the value of powerful AI models is increasingly bottlenecked not by model capability but by how well users can retrieve, organize, and build upon their accumulated interactions over time. Community feedback threads like this one on Reddit often serve as informal bug reports and feature requests that shape iterative product updates, suggesting this ambiguity around "Move to Project" may prompt clarification or a UI fix in future releases.
Read original article →