PUBLIC_KEY_PEM in the script before distributing to students.
There are four major AI tools you'll use throughout this course. This module is your first contact with two of them — and the beginning of understanding what these tools actually are.
By the end of this module, you will have:
That's it. Simple contact. No jargon required yet.
These are the four AI tools we work with in this course. Each comes from a different company, trained on different data, with different priorities. That diversity matters — one of the most important things you'll develop is a feel for which tool fits which task.
You don't need all four right now. You need two. Pick any two from the list below and sign up for free accounts.
Made by Anthropic. Exceptionally strong at reasoning through complex problems, working with long documents, and producing clear, well-structured writing. Tends toward careful, thorough responses.
Made by OpenAI. The original mainstream AI assistant — broadly capable, widely deployed, and the tool most of your colleagues have probably encountered first. The free tier runs GPT-4o.
Made by Google. Integrates with Google Workspace, has access to real-time web information, and handles multimodal inputs well. If your work lives in Google Drive, this one is worth exploring.
Made by a Chinese AI research lab. Open weights — the model itself is publicly available. Particularly strong at reasoning tasks. Provides a useful comparison point precisely because it was trained differently.
Copilot is a wrapper — its behaviour changes depending on whether you're in Teams, Office, Bing, or Windows. We study the underlying models, not the wrappers. If you already use Copilot at work, that knowledge applies, but Copilot itself is not part of this course.
Before you start, one thing worth understanding: these tools are not thinking. They're doing mathematics — very sophisticated mathematics — predicting what words should follow the words you gave them, based on patterns in an enormous amount of text.
That's important, because your brain is wired to socialize. When you're in a chat window typing to something that responds in natural language, your instinct is to treat it like a person. It isn't one. Understanding that changes how you interact with it — and how you get better results.
They can speak in plain language, but they do not think like humans. They are predicting words based on patterns in enormous amounts of text. This is exactly why your instinct is to treat them like a person — they respond in conversational language, they acknowledge what you say, they even apologize. But understanding that it's pattern-matching, not reasoning in the way you do, changes how you interact — and how you get better results from them.
The tool responds to what you give it. Every time you send a message, everything in the conversation goes in together. It starts fresh each time — with the full history you've built. What you give it determines what you get back. That's the whole game, and this course teaches you how to play it deliberately.
These tools do not remember past conversations. Every time you open a new chat, the tool has no idea who you are or what you've discussed before. If context matters to what you're asking, you have to provide it. This becomes the foundation of everything you learn from Module 2 onward.
Sign up for two of the four tools above. Then, in each one, send the message below exactly as written — with your name filled in. Don't modify it further. The point is to send the same message to both tools so you can compare the responses directly.
Sign up for free accounts on each. You'll need an email address. The free tiers are sufficient for everything in this course.
Start a new chat — not an existing one. We want a clean context.
Copy it, fill in your name, and send it. Then copy the full response you receive — you'll paste it into the form below.
Fresh conversation, same message, same process. Copy that response too.
Paste everything in. Hit Save. A file will download — send that file for your records.
In this module, we say send a message — not "write a prompt." Prompt is jargon, and jargon gets in the way. You'll learn what a prompt actually is in Module 2. For now: you're just talking to a tool. Use your own words.
Three fields. The message you sent, and the response you got from each tool. Fill them all in, then hit Save.
Paste the exact message you sent — with your name filled in. This should be the same text you sent to both tools.
Select which tool this is, then paste its full response below.
Select which tool this is, then paste its full response below.
When you click Save, your responses are saved to this file. A file will download automatically. Keep this file for your offline this module. Keep a copy for your records.