đź’Ľ Monday Business

Your First Week Using AI as a Project Manager

April 13, 2026

A practical day-by-day guide to getting started with AI tools—no hype, just useful first steps.

If you’ve been curious about AI but haven’t tried it for actual work, this is your starting point. Not theory. Not predictions about the future. Just a practical week of experiments that will show you where AI helps and where it doesn’t.

The goal isn’t to transform how you work. The goal is to try a few things, see what’s useful, and build from there.

Before You Start

Pick one tool. Don’t try to evaluate five AI tools at once. Pick one and use it for the week:

  • ChatGPT (free tier available) - The most widely known, good general-purpose tool
  • Claude (free tier available) - Strong at longer documents and nuanced writing
  • Microsoft Copilot - Integrates with Office tools

Any of these will work for this week’s experiments. You can explore others later.

Create a work-appropriate account. If your organization has an enterprise AI agreement, use that. If you’re using a personal account, don’t paste confidential information into it. For this week’s exercises, you can use anonymized or generic examples.

Day 1: Meeting Notes

The experiment: Take notes from a recent meeting and ask AI to structure them.

What to do:

  1. After a meeting, type or paste your rough notes into AI. You can also take a picture of handwritten notes—AI can read them.
  2. Use this prompt: “Turn these meeting notes into a structured summary with: key discussion points, decisions made, action items with owners, and open questions for follow-up.”
  3. Review the output. Edit as needed.

What you’ll learn: AI is good at imposing structure on messy information. Your rough notes become organized documentation in seconds instead of minutes.

Watch for: AI may invent details that weren’t in your notes. Always review before sharing.

Time investment: 10-15 minutes

Day 2: Prepare for a Difficult Conversation

The experiment: Use AI to help you anticipate questions and prepare responses for an upcoming conversation.

What to do:

  1. Identify a conversation you’re preparing for—a status update with a skeptical stakeholder, a budget discussion, a scope negotiation
  2. Describe the situation to AI: “I need to tell my project sponsor that we’re two weeks behind schedule due to a vendor delay. What questions will they likely ask, and how should I address them?”
  3. Review the suggested questions and draft your responses

What you’ll learn: AI is good at generating the questions you might not have thought of. It helps you walk into difficult conversations better prepared.

Watch for: AI doesn’t know your specific stakeholder or organization. Filter suggestions through your own judgment about what’s actually relevant.

Time investment: 15-20 minutes

Day 3: Draft an Email

The experiment: Use AI to help draft or improve an email.

What to do:

  1. Pick an email you need to write—a status update, a meeting request, a follow-up, or a response to a stakeholder question
  2. Either draft it yourself and ask AI to improve it, or provide the key points and ask AI to draft it
  3. Use prompts like: “Reword this email to be more concise” or “Draft an email to [recipient] about [topic]. Key points: [your bullets]. Keep it professional and to the point.”
  4. Review and revise the output in your voice

What you’ll learn: AI is good at restructuring your thoughts and finding clearer ways to say things. Email is low-stakes—you can experiment without much risk.

Watch for: The draft may sound generic or overly formal. Adjust the tone to match how you actually communicate. Never send without reading it first.

Time investment: 5-10 minutes per email

Day 4: Brainstorm Risks

The experiment: Use AI to generate a comprehensive risk list for a project.

What to do:

  1. Pick a current or upcoming project
  2. Describe it to AI: “I’m managing a [type of project] for [context]. What risks should I consider? Organize them by category.”
  3. Review the list. Mark which risks are actually relevant to your situation.

What you’ll learn: AI is excellent at generating breadth—risks you might not have considered. But it doesn’t know which risks actually matter for your specific project, team, and organization.

The pattern: AI provides breadth. You provide depth. Together, you get a better risk register than either would produce alone.

Watch for: AI will include generic risks that don’t apply. That’s fine—filtering is faster than brainstorming from scratch.

Time investment: 15-20 minutes

Day 5: Translate Technical Content

The experiment: Use AI to simplify technical content for a different audience.

What to do:

  1. Find a technical document, email, or specification you need to communicate to non-technical stakeholders
  2. Paste it into AI with this prompt: “Summarize this for a non-technical executive who cares about timeline, budget, and business impact. Remove jargon and keep it under one page.”
  3. Review the output for accuracy—make sure the simplification didn’t lose important nuances

What you’ll learn: AI is effective at adjusting complexity and tone for different audiences. Translation work that used to require multiple drafts happens faster.

Watch for: Simplification can lose precision. Technical stakeholders may need the original; executives need the translation. Know your audience.

Time investment: 15 minutes

Weekend Reflection: What Worked?

After five days of experiments, take 15 minutes to reflect:

What saved time?

  • Which experiments produced immediately useful output?
  • Where did you find yourself editing lightly vs. rewriting completely?

What didn’t work?

  • Which experiments produced output you couldn’t use?
  • Where did AI miss important context?

What will you keep doing?

  • Pick 1-2 uses that actually helped
  • Build those into your regular workflow
  • Ignore the rest for now

Week 2 and Beyond

If Week 1 went well, here’s how to expand:

Week 2: Use AI with meeting transcripts. If your conference call tool produces a transcript, feed it to AI for summaries, action items, or to clarify something you missed. This only works if your meetings generate transcripts—check your Teams, Zoom, or Google Meet settings.

Week 3: Try AI for more complex documents—a project charter, a business case outline, an RFP section. Remember to work in sections for longer documents.

Week 4: Experiment with brainstorming beyond risks—stakeholder concerns, alternative approaches, lessons learned questions.

Ongoing: Develop your own prompt library. When you find a prompt that works well, save it. Over time, you’ll build a personal toolkit.

If you have enterprise AI tools: Explore what’s possible with your organization’s data. For example, enterprise Copilot can create a stakeholder register from a list of employee names—give it the names and the columns you want, and it pulls role and department from your organization’s directory. You’ll still need to add context it can’t look up (influence level, communication preference), but the basic lookup work is done for you. Consumer AI tools can’t do this because they don’t have access to your organization’s directory. Check what integrations your enterprise tools offer.

Common Mistakes to Avoid

Mistake 1: Pasting confidential information into consumer AI tools. If your organization doesn’t have an enterprise AI agreement, assume anything you paste could be stored or used for training. Anonymize sensitive information or don’t use AI for that content.

Mistake 2: Using AI output without review. AI presents everything with equal confidence, including things it made up. Always review before sharing, especially for factual claims.

Mistake 3: Trying to use AI for everything. Some tasks are faster to just do yourself. AI is a tool for specific situations, not a replacement for thinking.

Mistake 4: Expecting AI to know your organization. AI doesn’t know your company’s politics, history, or unwritten rules. Apply your judgment to everything it suggests.

Mistake 5: Approving AI actions you don’t understand. When AI tools ask for permission to do something—run a command, edit a file, access something—make sure you understand what it wants to do before you approve. If you don’t understand, ask for clarification. You’re responsible for what happens.

The Bottom Line

Treat AI like an assistant or an intern—helpful, fast, but you still need to review their work.

This isn’t about becoming an AI expert. It’s about finding the 2-3 uses that actually make your work better and building those into your routine.

Most PMs who try AI find value in meeting notes, document drafts, and brainstorming. Some find value in preparation and translation. A few find it useful for analysis and synthesis.

Your job is to figure out which uses work for you. This week gives you a structured way to find out.

AI is a tool. Like any tool, it takes some experimentation to figure out how it fits into your work. Give it a week. See what helps. Build from there.