---
title: "Personal knowledge management with AI | Optimize All Academy"
description: "Your second brain, with a helpful assistant Personal knowledge management (PKM) means capturing, organizing and reusing what you learn: notes from…"
url: https://optimizeall.com/learn/ai-for-everyday-productivity/personal-knowledge-management
updated: 2026-10-05
---

AI for Everyday Work: Writing, Research, Meetings & Data · Translation and personal knowledge · lesson 17 of 17 · 12 min

# Personal knowledge management with AI

## Your second brain, with a helpful assistant

Personal knowledge management (PKM) means capturing, organizing and reusing what you learn: notes from meetings, articles, courses, ideas and templates. AI makes this much easier: it can tidy messy notes, tag and summarize them, and help you find and combine ideas later.

## A simple system: Capture, Organize, Use

**1. Capture quickly.** Use whatever is easiest: a notes app, voice memos, a document. Don't worry about neatness.

**2. Organize with AI.** Once a week, paste or upload your raw notes and ask:

```text
Here are my raw notes from this week.
1. Group them by topic.
2. Summarize each group in 2-3 bullets.
3. Pull out action items and ideas worth keeping.
4. Suggest 3-5 tags for each group.
Keep my wording where it matters; don't add information that isn't in my notes.
```

Save the result in your notes app or a well-organized folder.

**3. Use it.** When you need to write or decide, ask your assistant to work with your saved notes:

```text
Using my notes below on [topic], draft a one-page summary of what I've learned and three ideas I could apply to [project].
Tell me which note each idea comes from.
```

## 2026 tools for a second brain

- **Gemini Notebook (formerly NotebookLM):** add sources (docs, PDFs, slides, web pages, recordings) and ask questions answered only from them, with citations; generate study guides, mind maps and audio or video overviews. Each notebook now includes a secure cloud computer that can run code over your sources.
- **Notion AI:** search and summarize across your Notion workspace and connected apps, draft from your pages, and use AI Meeting Notes; Custom Agents can file and tag notes automatically on a schedule or trigger.
- **Projects in ChatGPT and Claude, Copilot Notebooks:** hold reference files and instructions for a topic, so each chat starts with that context.
- **Memory** in assistants remembers preferences and facts about you; review it regularly.

## Hands-on: build a topic notebook in 20 minutes

1. Pick one topic you must master this quarter (a client's industry, a new product, a certification).
2. Create a notebook (Gemini Notebook) or project and add 5 to 10 trusted sources: official docs, your meeting notes, key reports.
3. Run these prompts:

```text
Create a one-page briefing on [topic] from these sources only.
Structure: key concepts, current state, open questions, and a glossary of 10 terms.
Cite the source for every point.
```

```text
Quiz me with 8 questions on these sources, one at a time. After each answer, tell me
what I got right, what I missed, and the source passage to reread.
```

```text
What do these sources disagree about, and what important questions do they not answer?
```

4. Spot-check five citations by opening the passages.

**Before:** 14 browser bookmarks and a folder of PDFs you "will read later".

**After:** a cited one-page briefing, a glossary, a list of open questions to ask your manager, and a quiz that shows you which two concepts you haven't understood yet.

## Automate the weekly tidy-up (optional)

A Notion Custom Agent or a scheduled assistant task can run every Friday: "Group this week's notes by topic, extract action items into my task list, and add tags. Don't delete anything." Review what it did before relying on it.

## Where to go next

You now have a complete everyday toolkit. To go deeper: [Prompt Engineering Foundations](/learn/prompt-engineering-foundations) builds a larger pattern library; [Mastering ChatGPT](/learn/mastering-chatgpt) and [Mastering Claude](/learn/mastering-claude) cover each assistant's projects, agents, connectors and admin controls in depth; and [No-Code AI Automation](/learn/no-code-ai-automation-n8n-make-zapier) takes the simple automations from this course into Zapier, Make and n8n.

## Learning faster with AI

- **Explain it simply:** "Explain [concept] as if I'm new to it, then give me one real-world example."
- **Quiz yourself:** "Ask me 5 questions about my notes on [topic], one at a time, and give feedback."
- **Connect ideas:** "What links my notes on customer service and my notes on product design?"
- **Build templates:** turn repeated tasks into checklists you store and reuse.

## Worked example

Zara, a new account executive in Dubai, is overwhelmed by information in her first month: client briefs, product details, internal processes. Each Friday she pastes her week's notes into her work assistant's project and asks for the organize prompt. She builds a tidy knowledge base: one page per client, a glossary of internal terms, and a checklist for onboarding new clients. In month three, when a colleague asks how the reporting process works, she shares her checklist, which becomes the team's standard.

## Sample weekly output

A good weekly organization output looks like: "Client A (tags: onboarding, reporting): prefers weekly updates on Mondays; approval needed from marketing director for any budget change. Action: send the Q2 plan by the 14th. Idea: create a one-page reporting template." Short, tagged and actionable.

## A 15-minute weekly routine

1. Gather the week's notes (5 minutes).
2. Run the organize prompt and save the output (5 minutes).
3. Review action items and move them into your task list; delete notes you no longer need (5 minutes).

Consistency beats perfection: a small weekly habit builds a valuable knowledge base over a year.

## Privacy and hygiene

- Don't store passwords, financial details or sensitive personal data in notes you share with AI.
- Keep client notes in approved work tools only.
- Review memory and project files periodically; delete what's outdated.
- Remember that AI summaries of your notes can drop nuance; keep originals for important material.

## Pitfalls

- Capturing everything and never reviewing.
- Letting AI add "facts" to your notes that weren't there.
- Mixing personal and client knowledge in one place.

## Video lecture: Personal knowledge management with AI

Lecture coming soon · 11 chapters · about 8 minutes. Read the full transcript below.

1. A second brain with AI
2. Why it matters
3. The librarian analogy
4. Capture, organize, use
5. Example 1: the Friday tidy-up
6. Example 2: the certification notebook
7. Watch me do it
8. Automate the tidy-up
9. Privacy and hygiene
10. Common mistakes
11. Recap and try this now

## Lecture transcript

### A second brain with AI

How much of what you learned last month could you find right now? The great article. The client's preferences from that call. The idea you had in the car. For most of us, knowledge leaks away as fast as we collect it. Personal knowledge management, a second brain, is the fix, and AI makes it far easier than it used to be. In this lecture you'll learn a simple capture, organize, use system, the tools that make it work in twenty twenty six, and how to build a topic notebook that answers only from your sources. You'll see two examples and watch me build a notebook live.

### Why it matters

Why does it matter? Because knowledge work runs on memory, and memory is unreliable. We spend a surprising amount of time searching for things we already knew, re asking colleagues questions they already answered, and rebuilding documents that already existed. Good notes compound. A glossary you started in month one saves you time every week after. A checklist you tidied up once becomes the team's standard process. And with AI, the boring parts, tidying, tagging and summarizing, now take minutes instead of hours.

### The librarian analogy

Think of AI as a very fast librarian. You bring the books: your notes, articles, reports and recordings. The librarian catalogs them, tags them and, when you ask a question, finds the right page. The key rule, and it's the heart of this lesson, is that a good librarian answers from your shelf, not from their imagination. That's why we tell the AI to keep our wording, not to add facts, and to say which note each idea came from. Tools like Gemini Notebook, which used to be called NotebookLM, are built this way: they answer only from the sources you add, with citations.

### Capture, organize, use

Here's the system. Capture quickly. Whatever's easiest: a notes app, a voice memo, a quick document. Don't worry about neatness. Organize weekly, with AI. Paste or upload your week's raw notes and ask it to group them by topic, summarize each group, pull out action items and ideas, and suggest tags, keeping your wording and adding nothing that isn't there. Then use it. When you need to write or decide, ask your assistant to work from your notes and tell you which note each idea comes from. The tools can be as simple as a notes app plus an assistant, or as integrated as Notion AI, a project in ChatGPT or Claude, or a Gemini Notebook.

### Example 1: the Friday tidy-up

A simple example. Zara is a new account executive in Dubai, overwhelmed in her first month: client briefs, product details, internal processes. Every Friday she spends fifteen minutes. Five to gather the week's notes, five to run the organize prompt in her work assistant's project, and five to move action items into her task list and delete what she doesn't need. After three months she has a tidy knowledge base: one page per client, a glossary of internal terms, and a checklist for onboarding new clients. When a colleague asks how the reporting process works, she shares her checklist. It becomes the team's standard. Small, consistent, compounding.

### Example 2: the certification notebook

Now a realistic scenario. Tariq is an IT support lead in Birmingham preparing for a cloud certification while working full time. He creates a Gemini Notebook with eight trusted sources: the official exam guide, the vendor's documentation pages, his course slides and his own notes. He asks for a one page briefing from these sources only, with a citation for every point and a glossary of ten terms. Then he asks it to quiz him, one question at a time, and to point him to the passage to reread when he's wrong. The quiz shows two concepts he thought he understood but didn't. He spends his limited study time on exactly those. And because answers come only from his sources, he isn't learning from the internet's guesses.

### Watch me do it

Let me build one. My topic is a new client's industry, specialty coffee retail. I create a notebook and add seven sources: two industry reports, the client's brand guidelines, their last annual review, my kickoff call notes and two reputable trade articles. First prompt: create a one page briefing from these sources only, with key concepts, current state, open questions and a glossary. Cite every point. Second prompt: what do these sources disagree about, and what don't they answer? It points out that the two reports use different definitions of specialty. Useful to know before my next client call. Third, a quiz. Finally, my habit: I click five citations at random and read the passages. All five support the claims. Now I trust this notebook enough to use it.

### Automate the tidy-up

Once the weekly habit is working, you can automate part of it. In Notion, a Custom Agent can run every Friday afternoon, or right after an AI meeting note, to group new notes by topic, pull action items into your task database and add tags. In other assistants, a scheduled task can do something similar with the files you give it access to. Two rules. First, the automation should organize, not delete. Keep a human decision on anything that removes information. Second, review what it did for the first few weeks. Did it tag things sensibly? Did it miss an action item? Automation should make your fifteen minutes shorter, not replace your judgment about what matters.

### Privacy and hygiene

A few rules keep your second brain healthy. Keep work and personal knowledge separate, in separate tools or at least separate projects. Keep client notes in tools your organization approves. Don't store passwords, financial details or sensitive personal data in notes you share with AI. Review memory and project files every so often and delete what's outdated. And keep the originals of important material, because summaries drop nuance. The summary is your map. The original is still the territory.

### Common mistakes

Common mistakes. Capturing everything and never reviewing it, which creates a digital junk drawer. The weekly fifteen minutes is what turns notes into knowledge. Letting the AI add facts to your notes that weren't there. Always include keep my wording and don't add information that isn't in my notes, and spot check. And one giant notebook for everything, which makes answers muddled. Use one notebook or project per topic, client or course. And a final one: trusting a generated study guide or audio overview without checking. They're great for learning, but check a few citations before you share them with anyone.

### Recap and try this now

Let's recap. Capture fast and messy. Organize weekly with AI, keeping your wording and adding nothing new. Use your notes when you write and decide, and ask which note each idea came from. Use source grounded tools, like Gemini Notebook or a project with files, when you want answers only from trusted material, and check a few citations. Here's your try this now. Pick one topic you need to master this quarter. Build a notebook with five to ten trusted sources, run the briefing, quiz and disagreement prompts from the lesson text, and check five citations. Then put fifteen minutes in your calendar every Friday for the tidy up. In three months, you'll wonder how you worked without it.

## Key takeaways

- Capture quickly, organize weekly with AI, and use your notes when you write or decide.
- Source-grounded notebooks such as Gemini Notebook answer only from your sources, with citations.
- Ask the AI to keep your wording, not add facts, and say which note each idea came from.
- Keep work and personal knowledge separate, keep client notes in approved tools, and prune regularly.

## Try it

Collect a week of your notes, run the organize prompt, and save the result in a structured place. Then use it once to draft something useful.

- [Previous: Translation and localization](https://optimizeall.com/learn/ai-for-everyday-productivity/translation-and-localisation)
- [All lessons of AI for Everyday Work: Writing, Research, Meetings & Data](https://optimizeall.com/learn/ai-for-everyday-productivity)