Last updated
Learn what physically happens when AI runs. See which parts of a computer matter for AI, why GPUs and AI chips took over, why memory is the real limit on model size and speed, how quantization shrinks models, and what goes on inside the data centers that train and serve them. No code, no math.
Every AI answer is a physical process: electricity moving through chips and numbers moving between memory and processors. Understanding that layer explains many things people wonder about, like why AI runs on GPUs, why some models fit on a phone while others need a building full of servers, and why bigger models answer more slowly. This short course explains the hardware behind AI in plain language, with no code and no math notation. You will start with the parts of a computer that matter for AI and why moving data is often the real bottleneck. Then you will learn why GPUs and specialized AI chips took over, based on the kind of math neural networks do, and why memory, not raw speed, usually limits which model you can run and how fast it writes. You will estimate a model's memory needs with one simple multiplication. Next, you will see how quantization shrinks models by storing numbers with fewer bits, why models tolerate it, and where quality starts to drop. Finally, you will look inside AI data centers: how models are split across many chips, why the network between chips matters, how training differs from serving users, and why power, cooling, and water have become central to AI's growth.
2 modules • 5 lessons
Finish every lesson and pass the final exam to earn this free, shareable certificate.
Certificate of Completion
has successfully completed
5 lessons · Final exam passed
Michail Ouroumis
Founder, FreeAcademy.ai
Sample preview. Your name appears on the certificate when you complete the course. Learn more
Teaching a class or training a team? Assign this course for free
Neural networks mostly do huge numbers of simple multiply-and-add steps, and most of them can run at the same time. GPUs have thousands of simple cores built for exactly that kind of parallel work, so they handle AI math far faster than general-purpose processors.
No. The course explains everything in plain language with everyday comparisons. There is no code and no math notation, only one simple multiplication to estimate memory.
It gives you the concepts, like memory, bandwidth, and quantization, that decide what a device can run. For a hands-on guide to running models on your own computer, see Run AI on Your Own Computer.
It explains the physical side: why AI uses a lot of power, how data centers are cooled, and why water use matters. The wider ethical debate is covered in AI Ethics & Responsible AI.
About 45 minutes across five short lessons. Each lesson has a quiz, and there is a final exam with a free certificate when you pass.
A short, no-code, no-math explainer of the attention mechanism that powers ChatGPT, Claude, and Gemini. Understand query, key, and value, why models use many attention heads, how word order is encoded, and why transformers read your whole prompt at once.
Understand the machine underneath modern AI without writing a line of code. Learn what a neuron does, why weights are the only place a model stores knowledge, how layers build their own concepts, and how training actually works through loss, gradient descent, and backpropagation.
Run a free, private ChatGPT-style AI model right on your own laptop with Ollama and LM Studio. A beginner micro course: understand why local AI is private and free, check what your computer can actually run, install both tools, choose the right model, and use it every day to chat, draft, and summarize documents. No coding background needed, and your data never leaves your machine.
Some AI models pause to "think" before they answer. Learn what reasoning models actually do, why thinking step by step helps, how models are trained to reason, when a reasoning model is worth the extra time and cost, how to prompt one well, and where their thinking still goes wrong. No code, no math.
Understand what really happens when an AI agent books, searches, codes, or browses for you. Learn how a chatbot becomes an agent, follow the agent loop step by step through a real trace, see how tools and connectors work, why agents forget, why they fail, and how to judge an agent product before you trust it. No code, no math.
Find out why ChatGPT, Claude, and Gemini feel helpful, polite, and safe instead of just completing text. A short, no-code course on reinforcement learning, supervised fine-tuning, reward models, PPO and DPO, and the side effects of alignment like sycophancy and reward hacking. The natural next step after How LLMs Actually Work.
