--- title: "Use AI For This: Practical Playbooks for Real Life" date: "2026-06-15" author: "Graham" description: "A plain-English library of AI playbooks for everyday tasks: emails, quotes, studying, medical bills, small business workflows, and better follow-up prompts." tags: ["Playbooks", "Power of AI"] url: "https://powerofai.ca/use-ai-for-this" readTime: "7 min" --- # Use AI For This: Practical Playbooks for Real Life Most people do not need a lecture about artificial intelligence. They need one good example that makes the light turn on. This is the practical side of Power of AI: choose a real task, give the AI the right context, ask for a first draft, then push it with follow-up questions until it becomes useful. Not magic. Leverage. ## Quick picks - **Best first habit: Give context before asking.** The model is trained on the world, but your task lives in your details. Paste the email, quote, notes, rubric, policy, or document. - **Best second habit: Ask for a follow-up.** The first answer is usually the start. The second and third prompt are where the work gets sharper. - **Best safety habit: Make it list assumptions.** If the AI has to name what it is guessing, you can spot weak answers faster. - **Best model habit: Try another top model.** ChatGPT, Claude, Gemini, and NotebookLM feel different. You only learn the difference by using them. ## The simple playbook pattern Every useful AI workflow has the same shape: explain the goal, paste the context, ask for the output format, check the weak spots, then ask a follow-up. If you only ask one vague question, you are using a powerful tool like a search box. The better move is to treat the AI like a patient assistant. Give it the messy material. Tell it what good looks like. Ask it to show its assumptions. Then make it revise. - Goal: what you are trying to accomplish. - Context: the facts, files, notes, emails, quotes, rules, or constraints. - Output: the format you want back. - Check: the mistakes it should watch for. - Follow-up: the second prompt that makes the answer better. ## Why this matters now A lot of people tried ChatGPT in late 2022, saw one wrong answer, and put the whole category in the mental junk drawer. That was understandable then. It is a bad read now. The models are stronger, the apps can work with files and images, search is more connected, context windows are larger, and coding agents can inspect a real project instead of just guessing from a prompt. The old "it cannot count the letters in strawberry" test does not tell you what the best systems can do for real work today. ## Use AI where it multiplies judgment AI should not replace your judgment. It should get you to a better first draft, a better question, a cleaner plan, or a second set of eyes. The person who learns that loop becomes much more capable. The point is not to become less human. The point is to remove blank-page friction, compare options faster, explain things more clearly, and learn by asking better questions. - Use AI to draft, compare, summarize, organize, explain, and practice. - Use humans for final judgment, sensitive decisions, professional advice, and anything with real-world risk. - Use files and source material when the answer needs to be grounded. - Use top models when the task is hard enough to matter. ## Copyable prompts ### Universal use-ai-for-this prompt ```text I want to use AI for this task: [TASK]. My goal is [GOAL]. Here is the context: [CONTEXT]. Give me a useful first draft or plan. Then list assumptions, risks, missing information, and the best follow-up question I should ask next. ``` ### Second-pass improvement prompt ```text Improve your answer using this standard: [STANDARD]. Make it more specific, remove generic advice, show what you are assuming, and give me a version I can actually use today. ``` ## Related Power of AI pages - [Power of AI in 10 Minutes](/power-of-ai-in-10-minutes): The shortest path for someone who wants to understand what changed. - [You Tried ChatGPT Once. Try Again.](/you-tried-chatgpt-once-try-again): For people who bounced off early AI because it was wrong, weird, or underwhelming. - [Use AI to Write a Difficult Email](/use-ai-to-write-a-difficult-email): A practical first playbook almost everyone can use. - [Use AI to Compare Quotes](/use-ai-to-compare-quotes): Turn messy estimates into a clearer decision. - [Use AI to Help Your Kid Study](/use-ai-to-help-your-kid-study): Use AI as a tutor without turning it into a homework machine. - [AI Mistakes to Watch For](/ai-mistakes-to-watch-for): Learn the failure modes before trusting the output. ## Sources and official references - [Google people-first content guidance](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) - [Google AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) - [OpenAI ChatGPT capabilities overview](https://help.openai.com/en/articles/9260256-chatgpt-capabilities-overview) - [Anthropic Claude models overview](https://docs.anthropic.com/en/docs/about-claude/models) - [Google Gemini app evolution](https://blog.google/innovation-and-ai/products/gemini-app/next-evolution-gemini-app/)