# What is an AI second brain? > An AI second brain is one person's notes, kept where an AI tool can read them. What it does, how people build one, and where it stops for a team. Source: https://dienox.com/ai-second-brain Last updated: 2026-10-01 An AI second brain is one person's notes and knowledge, kept where an AI tool can read them, so the AI answers from what that person knows. It serves one person, in that person's own tools. A team needs the shared version: a company brain, with roles and a history of every change. ## What a second brain is. This page is about the personal kind: one person, their notes and their AI tool. For a team, read [What is a company brain?](https://dienox.com/company-brain) A second brain is a personal system. You keep notes, ideas and references outside your head, often in a note tool, organised the way you think. The name comes from Tiago Forte's [Building a Second Brain](https://www.buildingasecondbrain.com/). The idea is simple. Your head is for thinking. The system is for remembering. The hard part was never writing notes. It was finding the right one at the right moment, and keeping the system in order. ## What AI changes. An AI tool that can read your notes changes three things. - **It reads them.** All of them, when a question needs it. You no longer have to remember where something is. - **It answers from them.** Ask what you decided about a project, and the answer comes from what you wrote down, not from general knowledge. - **It can keep them in order.** File what is new. Link what belongs together. Point out where two notes disagree. The third goes furthest. In April 2026 Andrej Karpathy described [a pattern he calls an LLM wiki](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f): the AI model writes the pages itself and keeps them current from your sources, and you read. It is one way people build an AI second brain. [An LLM wiki for your whole company](https://dienox.com/llm-wiki) explains the pattern. ## How people build one today. You can build a second brain with AI from three parts. 1. **A place for your notes.** A note tool, or a folder of plain text. Somewhere that is yours. 2. **An AI tool.** Claude or ChatGPT, or a tool such as Claude Code or Cursor for those who like to build. 3. **A way in.** The AI tool has to reach the notes. Either you add them to a project in the tool, or the tool reads them where they live, through a connector. A connector is how an AI tool reaches something outside itself. Most AI tools use [MCP](https://modelcontextprotocol.io/) for that, the open standard for connecting AI tools to data and systems. A second brain that is connected through MCP can be read by any AI tool that speaks it: Claude today, another tool tomorrow. Start small. Put in what you look up most. Let the AI help keep it in order, and read what it writes. ## Where an AI second brain stops. For one person it works well. A team is not one person. What an organisation knows is spread over many heads, inboxes and documents. A personal second brain cannot hold that, and it was never meant to. - **It has one owner.** A colleague cannot read your notes, and their AI tool cannot either. - **It has no roles.** Nobody decides who may change what, because only you change anything. - **It has no shared history.** When a decision changes, nobody else sees what changed, or why. - **It lives in one place.** What a note tool holds, the AI tools of the rest of the team do not read. ## From one person's to a team's. At work the question changes. A second brain for work has to hold what the team knows, not what one person noted. Who decides what. What was decided. What you sell. How the work runs. Table: A personal AI second brain next to a company brain | | A personal AI second brain | A company brain | | --- | --- | --- | | **Who it is for** | One person | The whole organisation | | **Who reads it** | That person and their AI tool | People and every AI tool | | **Roles** | None: it has one owner | Yes: roles decide who may change what | | **History of changes** | Whatever the note tool keeps | Every change, and each can be undone | | **Who keeps it current** | Its owner | Everyone, as they work | The shared version has a name of its own: a company brain. [What is a company brain?](https://dienox.com/company-brain) compares it with a wiki, enterprise search and one AI tool's projects, and says what to put in first. ## Where Dienox comes in. Dienox is a platform that gives a company one knowledge base that every AI tool its people use reads from: Claude, ChatGPT, Copilot, Cursor and any other MCP client. It is for the organisation, not for one person's notes. Dienox calls it universal AI context, and the home page draws it as one picture: [every tool wired into one knowledge base](https://dienox.com#universal-ai-context). Your own second brain can stay where it is. It still helps you think. The company's knowledge base is for what everyone needs to know, and for every AI that works on the organisation's behalf. The [glossary](https://dienox.com/glossary#ai-second-brain) puts the two side by side in a few sentences each. ## Questions people ask. ### What is an AI second brain? An AI second brain is one person's notes and knowledge, kept where an AI tool can read them, so the AI answers from what that person knows. It is a second brain, the personal system for notes outside your head, with an AI as its reader. It serves one person. A team needs a company brain. ### What does an AI second brain do? It lets your AI tool answer from your own notes instead of from general knowledge alone. Ask what you decided about a project, and it looks in what you wrote down. It can also keep the notes in order: file what is new, link what belongs together and point out where two notes disagree. ### How do I build an AI second brain? You need three things: a place where your notes live, an AI tool, and a way for the tool to read the notes. That way is a project you add notes to, or a connector to the place they are kept. Start small, with the notes you look up most, and let the AI help keep them in order. ### Can I build a second brain with Claude or ChatGPT? Yes. Both can work from notes you give them: in a project, where you add them yourself, or through a connector to the place your notes live. The AI then answers from your notes and can help keep them in order. What you build this way is one person's, in that person's own tools. ### Is an AI second brain the same as a company brain? No. An AI second brain is personal: one owner, organised the way that person thinks, read by that person's AI tool. A company brain is shared. It holds what the whole organisation knows, the AI tool of every colleague reads it, roles decide who may change what, and every change is kept in its history. ### Can a team share an AI second brain? Not as it is. A personal second brain has one owner, no roles and no shared history, and the AI tools of colleagues do not read it. A team needs the shared version: one source that every AI tool reads, with roles and a history of every change. That is a company brain, and it is what Dienox provides. Sources - [Building a Second Brain, by Tiago Forte](https://www.buildingasecondbrain.com/) - [Andrej Karpathy: LLM Wiki](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f) - [What is the Model Context Protocol?](https://modelcontextprotocol.io/) The beta is open Be one of the first organisations on Dienox. We let organisations in a few at a time, so each one gets our full attention. Sign up in two minutes, and we come back to you within two working days. [Join the beta](https://dienox.com/beta)