Canada's AI literacy initiative, in brief
On June 4, 2026, the federal government launched AI for All, Canada's new national AI strategy. One of its measures: a National AI Literacy Initiative "that will offer entry-level AI training for all Canadians."
On September 9, 2026, the government launched that initiative with Amii, the Alberta Machine Intelligence Institute. The announcement describes streams for students, educators and Canadians in general, with different rollout dates. Read the government announcement for current access conditions.
This article isn't part of that initiative. It's a hands-on companion for one everyday situation: a conversation with an AI that runs long.
Habit 1: tell the AI who you are, once
Every new chat starts from zero about you. So you find yourself repeating the same things: "keep it short", "plain language", "I'm in Canada, use Canadian spelling", "give me options before you pick one".
Write those once, and put them where the AI reads them every time: in its settings (ChatGPT and other AIs have a place for instructions about you), or at the top of a new chat.
With Uncook: AI Profile reads your past conversations and writes a profile of how you like to work, based on what you actually asked for, with a short quote or count behind each point. Review it, then paste it into your AI's settings or at the top of a fresh chat. The new AI gets your preferences, not a memory of your old chats. AI Profile is free with an account, within its own reading and usage limits. It uses no tasks from your balance.
Habit 2: one chat, one job
A chat that starts with a recipe, moves to a cover letter and ends with a tax question gets confused, and so do you when you look for something later. Keep one chat per piece of work. When the subject changes, open a new chat.
Habit 3: know when a chat has gone too long
Long chats tend to get worse before they stop. Watch for:
- Slower answers, especially in the browser.
- Forgotten rules: the AI drops something you set early on.
- Going in circles: it suggests again what you already rejected.
- A hard stop: a message that the conversation has reached its maximum length.
None of these means you did something wrong. They mean it's time for a fresh chat. More on why: why long AI chats get worse.
Habit 4: start fresh with a handoff
Starting over doesn't mean starting from scratch. Give the new chat a short first message with:
- The goal, in one line.
- What's decided: the choices you don't want reopened.
- The rules you kept repeating.
- Where you stopped.
Pick up where we left off. Everything below is settled, so don't ask about it again. GOAL: [one line] DECIDED: - [decision] RULES TO KEEP: - [rule] WHERE WE ARE: [the latest version, or where we stopped] Continue from here. Be direct, no recap.
With Uncook: Handoff writes that message for you. On the AI's page, click the Uncook icon and pick Handoff. You get a prompt with your goal, what's decided, the rules you set and where you stopped. Copy it into a new chat. It works for chats in ChatGPT, Claude, Gemini, Perplexity, Mistral Vibe (ex-Le Chat), Gemini Notebook and Google AI Studio.
AI Profile or Handoff?
AI Profile is about you: how you like answers, in every chat. Handoff is about one piece of work: what's decided and where you stopped. Use the profile once, and a handoff each time a chat gets too long.
Habit 5: check what matters
An AI can be clear, confident and wrong. A fresh chat helps it stay on track, but it doesn't make every answer right. Before you rely on a date, a number, a rule or a name, check it against a source you trust. For anything about health, money or the law, ask a professional.
Questions people ask
Is this the government's AI literacy course? No. Canada's National AI Literacy Initiative has its own free courses, announced on canada.ca. This is a practical guide to one everyday skill.
Does a fresh chat make the AI more accurate? It helps it stay on track, because it's not re-reading a long, messy history. It doesn't guarantee the answers. Check what matters.
Will the new chat remember my old one? Not the thread itself. A handoff carries what you decided; an AI Profile carries how you like to work. Neither is a memory of every message.
Start fresh,
without starting over.
Add to Chrome
We don't keep the conversations you bring. When you click Send to Uncook in the extension window, the extension sends the chat from your tab to our servers, encrypted, for up to 30 minutes, and we delete it as soon as your Uncook page opens it. What Uncook makes from it is kept as a task, for as long as our Privacy page says. Neither your chats nor your tasks are used to train AI models.