Before you begin — about this guide

— Turning what is already on your computer into an asset —

What this guide is for

The first guide, Generative AI for Therapists: Foundations, covered how to meet generative AI, the differences between the major tools, what they are good and bad at, role setting, file uploads, and GPTs and Gems. Its aim was to get you touching the tools, and to give you a feel for AI as a set of intellectual training wheels.

This is the sequel, and its theme is

connecting the files on your computer to generative AI

A step beyond chatting in a browser or on a phone: learning to work with your local files — Word documents, PDFs, data, notes — alongside AI.

The truth is that people who have genuinely built generative AI into their work through their own files are, therapists or not, still a small minority. Which is exactly why crossing this line creates a decisive difference in what you can get done. This guide exists to help you cross it.

And throughout, the thing therapists care about most — personal information and security — is dealt with first, and most carefully. Once that sits right, you can move forward without unease.

What you should come away with

• A view that lifts your use of AI from "chat" to "workspace"
• A clear line on personal information and security, so you can handle work files without anxiety
• The ability to organise the files you already have into an environment you can share with AI
• A research assistant of your own, in Gemini Notebook
• The ability to hand Word files and PDFs to AI and summarise, rewrite and reorganise at will
• A safe, entirely local knowledge base grown in Markdown and Obsidian
• An understanding of agentic AI such as Antigravity, and how to handle it safely
• A route from "using AI" to "designing the work together with AI"

Who this is written for

• Therapists who have finished Foundations and want the next step
• People comfortable chatting with ChatGPT or Claude, but not yet making use of files
• People buried in papers, documents and data, losing time to organising it
• People hesitant to hand work files to AI because of privacy and security worries
• People who want to build a knowledge base of their own and accumulate learning as an asset
• People who eventually want automation and tool-building, but want the ground underneath it first

If you already drive your local environment freely with code and APIs, parts of this will feel basic. It is written as a bridge across to the minority side.

You do not need to have read Foundations; this guide is built to be followed from the beginning.

How to read it

• Read from Part 1 in order. This guide is designed to build upward from the ground.
• Each chapter ends with a "try this" and a summary. Please read with your hands actually moving on your own computer.
• The screens, features, limits and prices described are as of the time of writing (2026). AI moves quickly, so details will change. The underlying way of thinking changes far more slowly.
• Before handling any file containing personal information, read the rules in Chapter 2.

How it is organised

Five parts.

• Part 1 | Foundations — Why local files, now. And the security line that lets you work without unease.
• Part 2 | Organising — Tidying the files you have into an environment you can share with AI.
• Part 3 | Practice — Handing files to Gemini Notebook, ChatGPT and Claude, and working with them.
• Part 4 | Going further — Growing an entirely local knowledge base with Markdown and Obsidian.
• Part 5 | Ahead — The world where AI operates the computer itself (agents, Antigravity), and the map beyond.

Let us begin.