New Course
Your machine, your data
Local AI has become genuinely usable on an ordinary laptop, and also easier to get wrong now that every app has a cloud button. This course gets you set up and shows you how to keep it private.
What You'll Learn
3 parts
From why local AI makes sense to a hybrid plan you can maintain.
Why and what
What local AI is, the trade-offs, and whether your computer can run it.
Getting started
Choosing an app and a first model, and chatting privately with your documents.
Private, safe and useful
Privacy checks, local models with other tools, maintenance and when to use the cloud.
Course Includes
Concept boardsFour short ideas per lesson
FlashcardsKey terms to review
Practice questionsReal work scenarios with hints
Linked resourcesGuides, prompts and official docs
Timed final exam8 questions, 12 minutes
Printable certificateUnlocked by passing the exam
Ready To Start
Begin with Lesson 1
Start with what running AI locally means, and the honest trade-offs against cloud chatbots.
Related
Before or after this course
Comparing apps? Read Ollama vs LM Studio vs Jan in 2026. Ready to set one up on Windows? Follow our private AI assistant guide. Planning to run an agent on your own hardware? AI Agents for Everyday Use covers permissions and limits.
Privacy & Storage
Course progress, exam results and certificate details are stored in this browser on this device. They may be removed if you clear browser data.
All Lessons
Course Map
0 of 8 lessons complete
ExamNot attempted
Resume atLesson 1
CertificateLocked
Module 1 — Why and What10 min
Lesson 1: Why Run AI on Your Own Computer?
A local model runs entirely on your laptop or desktop. Your words and files never leave the machine, it works offline and there is no monthly bill. The trade-off is that it is usually less capable than the best cloud models, and you look after it yourself.
Module 1 — Why and What12 min
Lesson 2: Will It Run? Memory, Graphics Cards and Model Size
The question that decides everything is how much memory you have. Model size, measured in billions of parameters, and how heavily it has been compressed determine whether it fits and how fast it runs.
Module 2 — Getting Started12 min
Lesson 3: Choosing an App: Ollama, LM Studio or Jan
Three free apps cover most people's needs. They all run the same kinds of model; the difference is the interface and what they connect to. In 2026 all three also offer cloud models, so part of choosing is knowing where that switch is.
Module 2 — Getting Started12 min
Lesson 4: Choosing a First Model
There are thousands of open models. You only need one or two. Pick a well-known family, a size that fits your memory, and a version made for chat, then try it on your real tasks.
Module 2 — Getting Started12 min
Lesson 5: Chatting with Your Own Documents, Privately
The most common reason to run AI locally is to ask questions about files you can't upload anywhere: contracts, medical letters, client records, internal reports. Here's how document chat works on your own machine, and where it falls short.
Module 3 — Private, Safe and Useful10 min
Lesson 6: When Local Isn't Local: Privacy Checks That Matter
Running a model on your computer doesn't automatically make everything private. Cloud model options, web search, tool connections, sharing a local server and ordinary app updates can all send data out. Five checks cover nearly all of it.
Module 3 — Private, Safe and Useful12 min
Lesson 7: Local Models with Other Tools: Coding Assistants and Agents
Local models aren't just for chat. Coding assistants, agents and even Claude Desktop can be pointed at a model on your own machine. That saves money and keeps code private, with the usual agent cautions and some quality trade-offs.
Module 3 — Private, Safe and Useful10 min
Lesson 8: Looking After It, and When to Use the Cloud Instead
A local set-up needs a little care: updates, disk space and the occasional new model. Finish with a simple plan for which jobs stay local and which go to the cloud, so you get privacy where it matters and capability where you need it.