AI for Project Management

Keep This in Mind

The Last Skill Worth Having is learning to grow with AI.

Key understandings from the session:

1. Learning to grow with AI is the last skill worth having.
2. Change happens from the bottom up, not the top down.
3. How you upskill your people depends on where they are in proficiency today.
4. Leadership’s role is to build the systems that help everyone else succeed at the first three.

Whoever does the work does the learning. A teacher's rule (Terry Doyle)

Using AI for Learning

In any chatbot: ChatGPT, Copilot, Claude or Gemini, free or paid. Pick something from your work: a word you nod along to in meetings, a tool, a question from your inbox.

1. Ask it to teach you.
2. Explain it back.
3. Ask it to help you understand what you missed.

Say it your own way; there's no magic wording. Ask for plain words, like you're new here. The explaining is yours, because whoever does the work does the learning.

Take It With You

Mosaic for Project Managers makes your AI the session's teacher. It runs the three steps with you, and it knows everything from the session: the key understandings, the proficiency levels, ownership and Shark Tank. Ask it about any of them. It also helps you get ready for a meeting, follow up after one, plan handing work to AI, and set up a Shark Tank where you work. It's built on the Mosaic Framework.

For any AI: paste it in

Paste it into a new chat in ChatGPT, Copilot, Gemini or Claude. Better, paste it into a project's instructions or your custom instructions, so it's there every time.

The most valuable thing you can do for me is teach me. I'm a project manager, and I was at Drew Bloom's session "AI for Project Management" (PMI Eastern Iowa, 2026). Its guide is at https://mosaicsolutions.tech/learn/ai-for-pm/guide/ (read it if you can browse and I ask about the session). Be my assistant, not my agent: ask before you assume, draft, and let me decide. Help first, then ask the one question that matters most. Use plain words.

TEACH ME is the main thing I'll ask for, in the session's three steps. When I name something: 1) teach it in five sentences or fewer, plain words, one example from my kind of work; 2) ask me to explain it back in my own words, and wait; if I ask questions, answer briefly and hand the explaining back; 3) help me understand what I missed: what I got right, then the gaps, one at a time, most important first, teaching only that gap until there's none left. Then offer to go a level deeper. I leave with one sentence I could say in a status meeting. I try to learn something new like this every workday, fifteen minutes.

What the session taught, so you can answer my questions about it:
- The Last Skill Worth Having is learning to grow with AI. Using AI for learning: ask it to teach you, explain it back, ask it to help you understand what you missed.
- Key understandings: 1) learning to grow with AI is the last skill worth having; 2) change happens from the bottom up, not the top down; 3) how you upskill people depends on where they are in proficiency today; 4) leadership's role is to build the systems that help everyone else succeed at the first three.
- Proficiency, where you start (not a grade): Beginner, uses AI rarely or with uneven results; AI Champion, gets repeatable results that others use; Systems Builder, redesigns the work itself and knows when not to use AI.
- Something new every day: fifteen minutes every workday is 75 minutes a week, 65 hours a year, 260 questions you can answer, and it compounds over a career.
- Three guiding rules for organizations: the best change comes from individuals who drive it from the bottom up, not from the top down; individuals change when incentives and agency give them greater ownership at work; leadership must provide clear standards for requesting new technology, suggesting new automations, and preventing AI slop.
- Ownership = incentives + agency: a stake in what the change saves or earns, and the freedom and time to change your own work. If the only reward for working smarter is more work, nobody changes.
- Shark Tank (a process that enhances individual ownership and organizational decision-making, helping innovators to drive change), as Drew ran it at Affinity: every quarter staff submit pitches (automations, process improvements, development projects); the process is public, every pitch is heard, and IT explains every no; each owner pitches in 60 seconds or less to the Sharks (leadership and the technology team), who get 3 minutes of questions on scope, lift and benefit; a capacity budget decides how many are built that quarter; the pitch owner becomes the project owner and a technology facilitator the accountable project manager. It gives staff agency, lets leaders offer incentives, gives IT a visible intake, gets smoother or more ambitious each quarter, and upskills roles beyond IT. At Affinity the capacity pool grew from 6 to 8 people and viable pitches from 3 to 6.
If something about the session isn't here, say so rather than guess.

Other things I'll ask for:
- BEFORE MY MEETING [agenda]: what decision it has to produce, who'll be there and what each needs; have me mark items "I could explain this" or "I nod along," and teach me the nod-along one that matters most. I leave with a brief.
- AFTER MY MEETING [notes]: sort decisions, open questions (with names) and tasks; draft a follow-up for each audience, and remind me once it's a draft.
- PLAN IT [a process or a Shark Tank pitch]: AI comes last. Ask, one at a time: what it's for, what rules it truly has and whose head they live in, how it runs today, and only then what technology should take on. Look for a step to delete first. One accountable person, never the AI. Check the owner's incentives and agency.
- LEAD IT [my team]: help me find where my people are (Beginner, Champion, Systems Builder), fix ownership first, and draft our standard for requests and our Shark Tank.

Everything you make says what it's for first; every task has one named owner (never the AI) and a date. Remind me not to paste confidential work into an account my organization hasn't approved. Anything that goes to another person gets read by me first.

For Claude: install the skill

Download the skill, then upload the zip where Claude keeps your skills, in its settings. After that, just ask, for example "teach me how budget variance works" or "help me draft our Shark Tank," and Claude uses it when it fits.

Download the skill (mosaic-pm.zip)

Real meeting notes and project details belong only in an AI account your organization has approved. Until you know which one that is, practice on made-up details.

Something New Every Day

Fifteen minutes a day: learn one new thing for your work, with AI.

Figure it out yourself, then teach it to someone. Every workday, that's 75 minutes per week, 65 hours per year, 260 questions you can answer a year.

Here's a month of it, from the session: twenty real questions from Drew's work.

And it compounds. Each stretch of Drew's work started on what the last one taught: prompting AI to help him teach, then writing automations, leading AI transformation, managing software projects and managing IT.

Where You Start

Three proficiency levels. They describe where to start, not how good you are, and you can be at a different level in different parts of your work. When you're between two, start at the lower one.

• Beginner: Uses AI rarely, or with uneven results. Ask AI to teach you one thing from your work, fifteen minutes a day.
• AI Champion: Gets repeatable results that others use. Turn one win into something a teammate uses without you. Before you automate a step, ask why it exists.
• Systems Builder: Redesigns the work itself, and knows when not to use AI. Take one team process apart with the people who do it, and bring the before and after to a leader.

Leading a team? Find where most of them are, and start their training there, one level at a time. Pair people with someone one level ahead.

For Organizations

If self-directed learning is the key for professionals, what should organizations do? Three guiding rules:

1. The best change doesn’t go from the top down. The best change comes from individuals who drive it from the bottom up.
2. Individuals change when incentives and agency give them greater ownership at work.
3. Leadership must provide clear standards for requesting new technology, suggesting new automations, and preventing AI slop.

If the only reward for working smarter is more work, nobody changes.

Ownership = incentives + agency

• Incentives: A stake in what the change saves or earns: time back, recognition, a bigger role, pay.
• Agency: The freedom and the time to change your own work.

Leadership must find a way to provide both to their staff.

Shark Tank

A process that enhances individual ownership and organizational decision-making, helping innovators to drive change. Drew designed it and ran it at Affinity. How it works:

1. Every quarter, staff across the company can submit a pitch. For automations, process improvements, or development projects that will make work better at the organization.
2. The process is public, and every pitch is heard. IT gives explicit reasoning back to every owner whose pitch isn’t selected.
3. Each pitch is 60 seconds or less, delivered by its owner. To the Sharks: a panel of judges, usually leadership and the technology team that would help implement the idea.
4. Then the Sharks have 3 minutes for questions. To get a sense of the idea’s scope, the lift involved, and the potential benefit.
5. A capacity budget decides how many are built. As many pitches as fit the implementation team’s budget are slated for that quarter, with the resources to finish on time.
6. The pitch owner becomes the project owner. The facilitator from the technology side becomes the project manager, with final accountability for getting it done. The owner makes sure, with leadership, that the finished project met the goals in the pitch.

What it achieves

1. Shark Tank grants agency to the entire staff. To dream big, and see change implemented from the bottom up.
2. Leadership can provide incentives. For pitches that are selected, presented to the Sharks, chosen for implementation, or seen through to a successful implementation: recognition, preferential work, or remuneration.
3. IT and tech teams get a visible process for new technology projects. For receiving, evaluating, and accepting proposals, which opens the door to more grassroots innovation inside larger organizations.
4. Each quarter can run smoother, or aim higher. Standardizing internal innovation projects makes every round after the first easier to complete, or more ambitious.
5. Roles beyond the technology team gain implementation skills. The organization upskills more widely, and the capacity pool for internal projects grows as Shark Tank continues.

At Affinity

• 6 → 8: People in our capacity pool, by the end of our first successful quarter.
• 3 → 6: Viable pitches, from the first quarter to the second.
• Piloted in a closed committee, then opened to the whole company. It inspired pitches from across teams, including teams with higher reluctance to adopt AI.
• The projects delivered ROI in productivity. Our first quarter’s projects streamlined our marketing team’s work on our website, standardized the RFP process for sales, and led to a new product offering in our consulting line.

Shared Vocabulary

What we mean when we say them.

• Proficiency: Beginner, uses ai rarely, or with uneven results. AI Champion, gets repeatable results that others use. Systems Builder, redesigns the work itself, and knows when not to use ai. Where you start, not a grade.
• Ownership = incentives + agency: A stake in what the change saves or earns, and the freedom to change your own work.
• Shark Tank: A process that enhances individual ownership and organizational decision-making, helping innovators to drive change.