Study the lesson
Work through every section at your own pace, from start to finish.
You have opened an AI chat tool. The cursor blinks, and you type something like write me an email or explain this to me. A response appears: technically coherent, rather generic, and not quite what you wanted. This is the most common experience people have when they first use a large language model: the AI is capable of remarkable things, but what it actually produces seems oddly flat. The usual conclusion is that the tool is overhyped. The more accurate conclusion is that the brief was thin. An AI language model works only from what you give it. There is no background knowledge about you, your organisation, your audience, or your intentions unless you supply it. Give it something vague and it will return something vague, confidently and fluently, and it will not tell you it did not have enough to go on. The good news is that effective prompting is not a technical skill, it is a communication skill: the same principles that make a good brief, a clear email, or a precise question also make a good prompt. This lesson gives you a practical framework for applying those principles, along with worked examples you can adapt immediately.
Unlock the full lesson, the quiz, and your verifiable completion.
Continue with GoogleBy continuing you agree to the Institute of AI terms of use and privacy policy.
It is explained in plain English and assumes no technical background. Anyone can start it today.
Section one is open to everyone. Create a free account to work through the rest and take the quiz.
A prompt is the input you provide to an AI language model: the text, question, or instruction the model uses as the basis for its response. Everything the model produces is a direct consequence of what it receives in that prompt. Understanding this requires one technical fact: large language models such as Claude, GPT, and Gemini have no inherent memory between separate conversations. Some chat tools now offer an optional memory feature that carries a few details across sessions, but by default each new session begins with a blank slate, and the model sees only what you have written in the current exchange. This is the context window, the totality of what the model can see at any one moment, and if you have not told it who you are, what you are trying to achieve, or who the audience is, then as far as the model is concerned those things do not exist. Think of the model as an exceptionally talented freelancer who arrived at your desk this morning. They are knowledgeable, well-read, and capable, but they know nothing about you, your organisation, your project, or your preferences. They have never met your clients and do not know your tone of voice. You would not hand this person a sticky note saying write something about the project and expect a polished result; you would give them a proper brief, covering who the audience is, what the goal is, what format the output should take, and what to avoid. The model is that freelancer, so brief it accordingly. There is also an important implication of how LLMs generate text: they are probabilistic systems, which means the same prompt can yield different outputs across separate runs. There is no single correct answer waiting to be unlocked by the right incantation. Prompting is a skill of clear thinking and clear instruction, not a search for a magic formula.
Work through every section at your own pace, from start to finish.
A short set of questions on what you have just covered.
Every completion has a certificate that you can share publicly for anyone to verify.
Pass the quiz and the Institute of AI issues you a record of completion with your score. Every completion has its own public verification page, so the link you put on your CV or LinkedIn profile can be checked by anyone, at any time.
Free lessons, a quiz to test what you have learned, and a completion you can verify publicly. All from the UK's professional body for artificial intelligence.
Every AI lesson is free to start, free to finish, and ends in a completion you can verify publicly.