Second Brain vs. GTD: Knowledge, Actions, and AI Context

A second brain helps you organize knowledge you can use again, while GTD, short for Getting Things Done, helps you clarify and manage commitments. Use the first to keep useful material and the second to decide what action it requires. You can combine the habits without putting every note on a task list.

For AI-assisted work, add a clear record of what people approved and where progress stopped. Tools like fenbs connect tasks with Decisions and rules and project context, so the next assistant can continue from the work rather than reinterpret a pile of notes.

What is each approach trying to organize?

The official GTD introduction↗ (opens in a new tab) describes five steps: capture, clarify, organize, reflect, and engage. Tiago Forte’s Building a Second Brain introduction↗ (opens in a new tab) describes a knowledge workflow for turning saved material into useful output. Both involve capture and review, but their immediate questions differ.

That distinction is an emphasis, not a rigid boundary. GTD includes reference material, and a second brain can support completing projects. You do not need a separate app for every concept. You need to distinguish something you might use from something you have committed to do.

For the general knowledge-system definition, see what a second brain is. This comparison focuses on the moment saved information becomes, or does not become, an action.

Which problem should you solve first?

Start with commitment management when you keep wondering what you forgot to do. A bookmarked article will not remind you to send the proposal you promised. Capture the commitment, clarify its next action, and keep the supporting article accessible from that action.

Start with knowledge organization when you know what to do but repeatedly reconstruct the background. You may remember to prepare a workshop yet lose the useful examples from last time. Saving and explaining those examples reduces the research you need to repeat.

A simple diagnostic is to ask what delayed the last three pieces of work. If the work disappeared from view, improve the action record. If the work was visible but the necessary material was scattered, improve the knowledge record. If you reopened an approved choice, improve the decision record.

Fix the most frequent failure with one small habit. Replacing your whole setup makes it difficult to tell which change helped. A reliable end-of-day review or a short source note can be more useful than a complicated workspace you have not learned to maintain.

How do you turn one captured item into useful work?

Say you are preparing a training session and receive an article about better examples. The article is reference material. “Revise the first exercise using a complete worked example” is an action. “The session teaches three common customer requests” is a project choice. Keep each understandable in its own right.

1. Capture the article with a sentence explaining why it matters to the session. Keep the source address and any limitations you noticed.
2. Ask whether it creates a commitment. If it is only interesting, leave it as reference; saving something does not have to add an obligation.
3. If action is needed, name the next visible step. “Training ideas” is a topic; “Draft one worked customer request for the first exercise” describes work someone can do.
4. Link the action to the relevant source and the current project choice. Include the question that still needs a person if the direction is unapproved.
5. After applying the idea, record the result. A useful observation can become a reusable lesson without leaving the completed action on your active list.

This is a suggested combination of the two approaches, not a requirement to classify every thought. A source can remain a source. Add structure when it changes what you will do or makes later reuse easier.

fenbs keeps tasks and project decisions together, helping a connected assistant distinguish the material behind an action from the instruction that governs it.

What extra context does an AI assistant need?

A good task title may tell a person enough because they remember the meeting behind it. A new assistant does not share that memory. Give it the desired result, the relevant sources, the decisions that apply, and the boundary of its authority.

For the training session, state the intended audience and the approved exercise format. Tell the assistant whether it may draft an example or whether it must ask before changing the session structure. If a choice remains open, write that explicitly rather than leaving the assistant to infer approval from a polished draft.

Ask the assistant to explain its next step before a meaningful change. It should be able to say which action it is doing and which decision it is following. If it offers a new direction instead, you can check whether the project record is missing a choice or whether it simply failed to read it.

The broader AI second brain guide covers reusable knowledge. In this workflow, reusable knowledge supplies understanding, while the work record supplies the current assignment. Neither should require copying an entire conversation into every new chat.

Keep personal commitments separate from a team project when other members do not need to see them. Shared context should contain the work information people are allowed to use, not every private note that happened to lead to it.

How should you review knowledge and commitments together?

During a regular review, inspect active commitments before browsing saved material. Check whether each project has a meaningful next action, whether anything is waiting on a person, and whether completed work has an accurate stopping point. Then ask which knowledge would help move those actions forward.

That order limits a common distraction: reading your library can feel productive while the work that depends on it remains undefined. Give each useful reference a destination when one exists. If no active work needs it, keep it as reference without inventing a task just to justify saving it.

• For an active task: Is the next step clear enough that another person could start?
• For a waiting task: What answer or event is needed, and who can provide it?
• For a reference: Why did I save it, and does it still support the claim?
• For a completed task: What was checked, and what useful lesson should survive?

Review the difference between finished drafting and finished work. An assistant might have written the exercise while a person still needs to approve it. Record both facts. Calling the whole task done too early makes the next session trust an outcome that has not happened.

How to do this with fenbs

1. Create a free account, then add an active project with a short description of the outcome.
2. Add the tasks needed for that outcome. Use clear action titles, include the relevant background, and keep source references in Links and listings.
3. Record the choices people made in Decisions and rules. Keep continuing instructions as rules and leave unanswered choices open.
4. Connect a supported assistant through Integrations. Ask it to read the project context before acting and work within the permissions you approved.
5. At the end of the session, save completed work, checks, remaining questions, and the next step in Where we left off. Review the record before the next session starts.

fenbs does not implement every GTD list or replace your calendar and personal reference library. Its kanban board uses To Do, Next Up, In Progress, and Completed. It is a good fit when your tasks need shared decisions and AI context; a simple personal list may be enough for private reminders.

What is a sensible first-week experiment?

Choose a project with a small deliverable, such as an internal guide or workshop exercise. Use your existing note tool for references and one clear record for actions and choices. Avoid moving your entire archive while testing whether the workflow helps.

After a few sessions, start fresh and ask a connected assistant to identify the next action, explain the approved direction, and name the reference it needs. Check that answer against the record. A useful setup should let you resume without retelling the meeting that created the task.

Explore other organizational approaches in the second brain collection, or read what an AI brain is for the broader work-context model. Keep the habits that make your next action clearer and your useful knowledge easier to apply.