Maple Launches Projects, Shared Context, and Pinning Features to Organize Private AI Workspace
Key Takeaways
- ▸Projects feature allows users to organize related chats into themed groups, with custom instructions that apply across all conversations within a project
- ▸Maple maintains its privacy-first approach—all projects and chats remain end-to-end encrypted on the user's device and are never used to train other models
- ▸Pinning feature and improved chat sorting (by recent activity) reduce friction in accessing frequently-used conversations
Summary
Maple has released a major update introducing Projects, a long-requested feature that allows users to organize their AI conversations into thematic groups while maintaining end-to-end encryption. The new feature addresses a common pain point where chat histories become unwieldy as conversations mix work, personal, and various other topics together. Users can now group related chats into Projects, pin frequently-accessed conversations, and set custom instructions for each project—all while keeping data encrypted on their device and processed within Trusted Execution Environments (TEEs).
The Projects feature transforms Maple from a one-shot query tool into a persistent workspace. Each project can have its own custom instructions that automatically apply to every chat within that project, eliminating the need to repeat context in individual conversations. Use cases include long-running work like dissertations, legal matters, client projects, courses, and specialized roles. Additionally, chats are now sorted by most recent activity, and users can pin their most frequently-referenced conversations to a dedicated sidebar section for quick access.
- Custom instructions per project enable users to set role context, client preferences, writing styles, language learning goals, and other persistent parameters without repetition
Editorial Opinion
Maple's Projects release represents a thoughtful evolution toward AI tools that function as true workspaces rather than disposable chat interfaces. By combining organizational features users already expect from competitors with a genuine privacy guarantee—where data remains encrypted and never fuels training pipelines—Maple is making a strong case for privacy-conscious professionals who need powerful AI capabilities without the surveillance trade-off. The custom instructions per project is particularly clever, as it acknowledges that most AI work requires consistent context and solves it without compromising user privacy.
