Artificial or Project Management Intelligence?

Ever since ChatGPT came out, I’ve been looking for ways to boost my project management work.

I was enthusiastic and open to the new tech.

So, I tried it, experimented with it, and soon I was using it heavily for my writing and content creation —training scripts, articles, all kinds of learning materials.

But for actual project management work?

I wasn’t engaging with it deeply.

Was I change-resistant?

Or not up to speed with the latest in AI?

That wasn't the case. One of my clients was a Silicon Valley-backed AI startup. I was getting insights about the latest in the AI field coming straight from the heart of the industry.

So it wasn't my attitude or knowledge – I was simply not finding strong use cases for project management work.

Meanwhile, my feed was full of bold claims:

"AI makes project managers 5–10× faster!"

"It automates tasks so we can focus on strategy.”

It sounded great, but it didn’t feel real to me.

Months passed. I kept reading the same promise again and again. I saw no big, tangible shift in my own projects—or in the work of respected PMs in my network.

This summer I decided to dig in again. I ran a lot of hands-on experiments across the daily PM work. I pushed myself and tried different prompts, different workflows, different ways of framing the same tasks.

And things started becoming clearer!

On one hand, there was a tangible jump in the technology. Chat-GPT, Copilot and few others started doing much more useful things that a PM can benefit from.

And so, after my latest “sprint” in the AI arena I want to share my latest thoughts and tips for project managers.

But first, let's understand one important point that is specific to our profession.

Finding #1: Big tasks VS Fragmented tasks

Project Management work is differently structured.

The World of Big tasks: Professions like engineering (software), marketing, and finance are more "stable" in terms of fewer, bigger tasks. In practice, that means you're more likely to find yourself working for few hours (or days) on the same task. It's big and important enough to keep you in front of the same window on your PC.

Those types of tasks are much more prone to efficiency gains from AI. A software developer can spend 4 hours straight coding a feature, feeding requirements to AI and iterating on solutions.

The Fragmented Reality: For PMs, however, it is different. Our work is broken into a great number of smaller steps, which are often randomly scattered around our work day. Our days are a chain of many small steps across people and tools:

status report → follow-ups → call → vendor ping → workshop prep → risk check → budget nudge → call again…

In those same 4 hours, I'll update 3 different status reports, join 2 calls, respond to 15 emails, escalate 1 issue, and prepare materials for tomorrow's workshop. Each task takes 15-30 minutes max.

Work that is small, varied, and spread out is harder to automate end-to-end. By definition, automation comes first at the door of "high-volume, low-complexity" tasks. Not the definition of our PM work.

With that in mind I have a question for you:

Why are PMs hired to manage projects? Is it mainly to make projects faster?

Finding #2: Don’t chase speed. Aim for quality.

The main goal with AI should not be SPEED. We should not be so simply looking to become faster.

Our main goal should be better project management—better plans, better documents, better communication, better decisions.

Saving a few hours is nice. But raising the quality of our outputs can change the whole project.

Example: AI can help you draft a project plan in hours, not days. Good. But the real win is a more comprehensive plan—one that surfaces hidden scope risks, highlights vague handoffs, and challenges weak assumptions. That plan reduces rework, prevents fires, and avoids painful escalations months later.

The value of that is not 2–3 hours saved. It can be 10x more PM value across the project.

That is where we should point AI: not at “go faster,” but at “think wider, go deeper, and help me make better decisions.”

Finding #3: Practice Real-life Use Cases for PMs

The level of usefulness is what we need to grasp. And let's face it - it's strongly correlated to your effectiveness as a manager - as an "AI manager" in that case.

Think of AI like you would any other resource: treat it like a stakeholder who can give you lots of info (e.g., senior expert) or like a team member, a PM assistant (perform specific tasks - draft docs, etc.).

This is a complex point for any professional as it requires that we make two skills come together without fighting - Artificial Intelligence and PM Intelligence.

The AI skills include your knowledge and practical skills when working with AI. But your project management intelligence will help you first decide when to use AI - for which PM tasks to use it and for which - not.

PM intelligence will also help us with managing bias. Our work is full of people, personalities, and emotions. PMs are not an exception. As AI can speed up and generate a lot from little, it can do that also with biases - amplifying a bias and making it worse for everyone.

Example: We may start a conversation with the feeling that Jim is blocking our way. The AI can easily pick up that and suggest a plan that is centered around battling Jim’s position. But what if Jim has a point? What if we were emotional? AI will amplify that and it can be in the wrong direction.

More senior PMs tend to have better perspective of all levels of the project, the circumstances, etc. They know in which light to present the project and task, as it will have a big influence towards the model's actions.

Project management intelligence is needed here.

Then Artificial Intelligence skills join the party. When you go into a specific use case for your PM work, you need to find the right way of prompting and strategizing your task.

• Are you going to do one full prompt?
• Are you going to use iterative prompting?
• Will you attach files or simply write?
• How will you define the exact output?
• Will you aim for final output to become a deliverable, or use the AI to produce useful "work in progress" for your project deliverable?

It all makes a difference. You need to keep learning and experimenting. If you had trainings 6 months ago you are behind again.

What's Next: My Practical Tips

Tip #1: Stop Following the Crazy News

Stop following the crazy news about AI's huge impact on the PM market. Focus on realistic, practical applications instead.

Tip #2: Focus on Quality Enhancement, Not Speed

Instead of asking "How can AI make me faster?" ask "How can AI make my work better?" Better risk analysis, more comprehensive planning, clearer communication.

Tip #3: Develop Your AI + PM Intelligence

This is a skill combination that requires practice. Start small, measure what actually works, and build from there.

Ok, I confess, I am working on a solution

After this summer, I started working on a new learning program to help PMs understand the use cases, learn AI capabilities that go beyond the simple ChatGPT window and, most importantly, do exercises to increase their AI fitness. Think of it as an AI gym for PMs.

If that sounds like something you might benefit from, message me and I'll add you to the waitlist. I promise it won't be another course telling you that AI will make you 10x faster - because we both know that's not how PM work actually works.

Remember: AI amplifies intelligence, it doesn't replace it. Your PM judgment is still the most critical component.