Perplexity Rebuilds Projects With Brain Memory to Beat Stateless AI Agents
Perplexity's Spaces become Projects: a persistent file system and self-improving Brain memory now power long-running agentic work for all users
- Spaces become Projects: Perplexity's Spaces are now Projects, with a persistent shared file system and self-improving memory for long-running agentic work.
- Persistent file system: Computer can now read, edit, and save files across sessions, so work builds on itself rather than starting over each time.
- Brain memory system: Projects integrate with Brain, which builds a context graph overnight from past sessions, corrections, and connector results.
- Measured gains: Perplexity's internal tests show +25% answer correctness on repeated tasks, +16% recall, and 13% lower cost on context-heavy workflows.
- Team-safe collaboration: Shared Projects keep files and context common, while personal memory and connector credentials stay scoped to each individual user.
- Pricing: Projects are available to all Computer users today; Computer requires the Max plan at $200/month (10,000 credits included).
Perplexity just rebranded and significantly upgraded Spaces into Projects, its new home for long-running work inside Computer, the company's agentic AI platform. The rename is the least interesting part. What's actually new is a persistent, shared file system and deep integration with Brain, Perplexity's self-improving memory system, turning Projects into something closer to a continuously learning team workspace than a simple chat organizer.
The problem with stateless agents
Every AI agent has the same fundamental flaw: it forgets. Start a new session and you're back to square one, re-explaining context, re-uploading files, re-specifying preferences. For one-off tasks, that's tolerable. For ongoing work spanning days, teams, and dozens of files, it's a real bottleneck. That's the gap Projects is designed to close.
Projects are built for how people really work. Each Project has a shared, persistent file system where humans and agents collaborate on the same files, and a Brain that learns from your work so the next task starts with full context from the last.
A file system that acts like a teammate
Computer works in the file system the way a teammate would. It can open files, read their latest versions, make edits, and save changes back to the same files. Computer edits the same file rather than starting over every task. That's a meaningful shift from the typical pattern where each agent run produces a new artifact you have to manually reconcile with the last one.
The file system is hierarchical and durable. You can:
- Upload documents, PDFs, decks, and spreadsheets directly
- Import local folders from the desktop app (Mac and Windows)
- Connect to over 400 external tools and apps
Users can connect Projects to over 400 tools and apps their team needs, including Google Drive, Notion, Linear, Snowflake, and GitHub. Local files stay private to the user who uploaded them, even inside a shared Project.
Brain: memory that learns from work, not just preferences
The more architecturally interesting piece is Brain. Most AI memory systems store things about you -- your name, your preferences, your writing style. Perplexity said the goal is to move AI memory beyond personalization. Most memory systems focus on the user, including preferences, contacts, work style, and recurring instructions. Brain is focused on what the agent did, what worked, what failed, and what corrections were made.