AI for Everyday Work

Lesson 1.1 — The One-Paragraph Version

AI like ChatGPT, Claude, and Gemini is a text-prediction engine. It has read an enormous amount of writing and learned the patterns of how words, ideas, and answers fit together. When you ask it something, it predicts the most useful response. It is not searching the internet by default. It is not “thinking” in the human sense. It is not always right. It is, however, extraordinarily good at language-based tasks—which is a large part of office work.

Lesson 1.2 — What It's Great At

• Rewriting and improving text
• Summarizing long documents
• Brainstorming options
• Explaining concepts at different levels
• Translating between formats, such as bullets → email or notes → report
• Drafting first versions of almost anything

Lesson 1.3 — What It's Bad At

• Recent facts unless current web access or trusted sources are available
• Precise calculations unless the tool is using a reliable calculator or code execution
• Anything requiring real-world verification
• Knowing your specific company context unless you provide it
• Reliably knowing when its own answer is wrong

Lesson 1.4 — The Golden Rule

AI is a first-draft machine, not a final-answer machine. You are always the editor. The people who get in trouble are the ones who forget this.

Lesson 2.1 — The Anatomy of a Good Prompt

Every good prompt has four parts:

1. Role — “You are an experienced project manager.”
2. Task — “Write a status update email.”
3. Context — “The project is 2 weeks behind due to a supplier delay.”
4. Format — “Keep it under 150 words, professional but warm tone.”

Lesson 2.2 — The Rewrite Loop

Don't accept the first draft. Ask: “Make it shorter.” “Make it warmer.” “Make it more direct.” “Rewrite for a skeptical audience.” Two or three loops can turn mediocre output into genuinely useful output.

Lesson 2.3 — Worked Example

Weak prompt: “Write an email to my boss.”

Strong prompt: “You are a marketing manager. Write a 3-paragraph email to my CEO explaining that our Q3 campaign underperformed by 15% because of a delayed launch, what we learned, and what we'll do differently in Q4. Professional, direct, no excuses, no jargon.”

The second prompt provides enough direction to generate a more usable first draft.

Lesson 3.1 — Feeding It Sources

Paste in articles, transcripts, PDFs, or meeting notes. Then ask specific questions: “What are the three main arguments?” “What did the author leave out?” “Summarize this for someone who has 30 seconds.”

Lesson 3.2 — The Brief in Five Minutes

Workflow: gather 5–10 sources → provide each to the AI → ask for a one-paragraph summary of each → then ask it to synthesize them into a single brief with key themes and open questions.

Lesson 3.3 — The Verification Habit

When AI gives you a fact based on supplied material, ask: “Which part of the source supports this?” If it cannot point to source evidence, treat the claim as unverified.

Lesson 4.1 — From Idea to Plan

Describe a goal in plain language. Ask the AI to break it into phases, tasks, owners, and a timeline. Then ask: “What could go wrong?” and “What am I missing?” AI can be a useful thinking partner for planning.

Lesson 4.2 — Meeting to Action

Paste in meeting notes or a transcript. Ask for decisions made, action items with owners, open questions, and a follow-up email.

Lesson 4.3 — The Pre-Mortem

Before starting, ask the AI: “Assume this project failed six months from now. What are the five most likely reasons?” This can surface risks you may have missed.

Lesson 5.1 — Uploading Files

Many modern AI tools let you upload PDFs, spreadsheets, Word documents, and images. Once uploaded, you can ask questions about them directly, subject to the tool's security and privacy settings.

Lesson 5.2 — Analyzing Data

Paste a table or upload a spreadsheet and ask: “What patterns do you see?” “What's the outlier?” “Summarize this in plain English for a non-technical reader.” Verify important calculations with a reliable spreadsheet, calculator, or code tool.

Lesson 5.3 — Building a Personal Knowledge Base

Keep a running document of your company's terminology, recurring tasks, and best prompts. Provide relevant context at the start of a session. AI becomes much more useful when it has the right working context.

Lesson 6.1 — What Never Goes In

• Customer personal data unless your organization has explicitly approved the tool and workflow for that data
• Confidential company information unless the platform and company policy explicitly permit it
• Legal, medical, or financial conclusions you will act on without qualified review
• Any sensitive information you are not authorized to disclose to that system

Lesson 6.2 — Check Your Company Policy

Many workplaces have rules about which AI tools are approved and what information may be used with them. Know your policy before using AI on work content.

Lesson 6.3 — Disclosure

If AI contributed materially to a report, proposal, or client deliverable, follow your workplace, client, professional, academic, or regulatory disclosure requirements.

Lesson 6.4 — The Human-in-the-Loop Rule

You are responsible for anything you send, publish, or act on. “The AI said so” is not a defense. Read the output, verify consequential facts, and own the final decision.