Feature request: Project-centric interface for heavy ChatGPT users

ChatGPT Feature requests
, ,
1

Feature request: Project-centric interface for heavy ChatGPT users

I use ChatGPT intensively across many simultaneous projects: writing, research, software ideas, teaching, publishing, business concepts, and long-term intellectual work.

The main limitation I encounter is not the quality of the model. It is the way the interface organizes ongoing work.

ChatGPT is still primarily organized around conversations. Projects improve this by grouping chats, files, and instructions, but for users managing many complex and interconnected initiatives, the conversation is often not the right primary unit.

What I need is a project-centric layer above chats.

A project should behave like a persistent object with its own state, not simply like a container of conversations.

For each project, ChatGPT could automatically maintain a structured overview such as:

  • objective;

  • current status;

  • estimated progress;

  • completed work;

  • pending tasks;

  • blocked tasks;

  • important decisions;

  • discarded ideas;

  • next recommended action;

  • relevant documents;

  • related projects;

  • dependencies;

  • last meaningful update.

The system should extract and update this information from normal conversations, so users do not have to maintain a second task-management system manually.

For example, after a long working session, ChatGPT could offer a “Close session” action and automatically identify:

  • what was completed;

  • what decisions were made;

  • what was rejected or postponed;

  • what remains unresolved;

  • what the next step should be.

The information would belong to the project rather than remaining buried inside the specific chat where it appeared.

A higher-level project dashboard

For users with many projects, there should also be a global view showing:

  • all active projects;

  • paused projects;

  • incubating ideas;

  • completed projects;

  • priority;

  • recent progress;

  • pending work;

  • projects that have not moved recently;

  • projects that depend on other projects;

  • projects that may belong to the same larger ecosystem.

The dashboard could answer a very practical question:

“Where should I put my next two hours?”

ChatGPT could suggest priorities using factors such as urgency, impact, effort remaining, current progress, dependencies, external opportunities, and the cost of leaving a project inactive.

Importantly, the system should distinguish between:

User priority — explicitly chosen by the user.

Suggested priority — calculated by ChatGPT.

The AI can recommend, but the user remains in control.

Projects are not always hierarchical

A traditional folder structure is not enough.

Many projects share knowledge, audiences, tools, files, or infrastructure. A graphical “constellation” or relationship view could show links such as:

  • depends on;

  • feeds into;

  • shares research with;

  • shares infrastructure with;

  • belongs to the same ecosystem;

  • could potentially be merged.

This would be especially useful for people whose work spans several disciplines or long-term creative projects.

Why existing task managers do not fully solve this

Tools such as ClickUp, Notion, Trello, and similar platforms are excellent at managing explicitly entered tasks.

The difference here is that the knowledge already exists inside ChatGPT conversations.

The opportunity is for ChatGPT to convert conversation into structured project memory automatically.

The user should not have to discuss a project with ChatGPT and then manually reproduce the same information in another productivity application.

The principle should be:

No double entry.

The underlying problem

I believe this becomes increasingly important as people move from using AI as an occasional assistant to using it as part of their daily intellectual and creative infrastructure.

For light use, organizing work by chat is sufficient.

For heavy use, conversations become fragments of larger processes.

The natural hierarchy becomes something closer to:

Ecosystem → Project → Subproject → Knowledge → Decisions → Tasks → Documents → Conversations

rather than:

Chat → Messages → Files

The chat is still essential, but it becomes one way of interacting with a persistent project rather than the place where the project’s state lives.

Possible interface

A project workspace could have several views:

Chat
Normal conversation with ChatGPT.

Overview
Objective, current state, progress, last update, and next action.

Tasks
Automatically extracted completed and pending work.

Decisions
A persistent record of meaningful decisions and rejected alternatives.

Files
Relevant documents and generated artifacts.

Timeline
Major changes and progress over time.

Relations
Connections and dependencies with other projects.

At the account level, a global dashboard could provide:

Now | Active | Paused | Incubator | Completed

along with a global search across projects, tasks, decisions, files, and concepts.

Why I think this matters

ChatGPT already helps users think, create, research, and produce.

The next step may be helping them preserve continuity across that work.

The problem is not simply memory. It is structured continuity.

A user with dozens of active conversations should not need to remember where a decision was made, whether an idea was abandoned three weeks ago, which project depends on another, or what the next meaningful action should be.

ChatGPT already understands much of this information during the conversation.

The feature I am suggesting is a way to make that understanding persistent, visible, editable, and useful across time.

I would personally use this every day.

1 Like