What’s Actually on Leaders’ Minds About AI + Project Execution?
Recently, I had the opportunity to talk about AI and project execution with two different Vistage groups: one made up of CEOs and another of key executives.
What happens to humans when AI saves us time?
We can spend human time where it matters.
Rather than spend all our time talking about what AI might eventually do, I asked them to think about their own businesses:
Where could AI and technology make project execution meaningfully better right now?
The ideas came quickly. And while the groups approached the question from different perspectives, several clear themes emerged.
1. Get Humans Out of Project Administration
This was probably the lowest-hanging fruit.
Project managers and team members still spend an extraordinary amount of time maintaining the machinery around a project: updating tasks, collecting status, scheduling meetings, taking notes, sending reminders, creating reports, rebuilding information in different systems, and chasing people for updates.
The opportunity isn't necessarily to eliminate the project manager.
It's to eliminate a lot of the work we make the project manager do.
If AI can automatically capture commitments, update status, organize project information, prepare reports, identify missing information, and tee up the conversations that require human attention, the project manager can spend considerably more time actually managing the project.
2. See Problems Before They Become Status-Report Problems
Another recurring idea was using AI to identify risk and slippage earlier.
Most traditional project reporting tells us what has already happened.
A milestone was missed. A task is late. The budget is over. The project has turned yellow—or worse, red.
But what if AI could continuously look at the signals underneath those outcomes?
Are commitments starting to slip? Is someone's workload making the original timeline unrealistic? Is actual performance beginning to diverge from the estimate? Is a critical dependency becoming a problem?
Even more useful: Why?
If someone missed a project commitment because three other business priorities consumed their week, that is different from simply knowing the task is overdue.
The opportunity is to move from reporting project history to anticipating project outcomes.
3. Make Better Decisions About What to Do in the First Place
This may be the bigger opportunity.
One group took the discussion upstream from project management entirely.
Before we use AI to execute a project faster, should we be asking whether it is the right project?
AI can potentially help leaders evaluate competing initiatives against strategy, expected ROI, available resources, existing commitments, historical results, and current business conditions.
It can also identify what we don't know.
Instead of confidently producing an answer from incomplete information, a well-designed AI-enabled process can ask:
What information is missing before we should make this decision?
That matters because executing the wrong project more efficiently is not exactly a win.
4. Use the Knowledge the Company Already Has
Companies have enormous amounts of institutional knowledge trapped in old project plans, bids, estimates, emails, meeting notes, spreadsheets, systems, and—most frustratingly—people's heads.
AI creates an opportunity to actually use it.
What did a similar project cost last time?
Where did the estimate go wrong?
What risks appeared?
Which tasks are normally required?
How long did it really take?
What was different about the successful projects?
Instead of starting every initiative with a blank sheet of paper, organizations can increasingly use their own history as context for the next decision.
5. Give Humans Better Information—not Just Less Work
This was the idea I found most interesting.
The real opportunity for AI may not ultimately be automation.
It may be better human decision-making.
Imagine identifying the important recurring decisions in a business, understanding the financial impact of those decisions, and then asking:
What information would allow this person to make this decision better—and can AI put that information in front of them at exactly the right time?
That's a very different AI strategy from simply asking, “What tasks can we automate?”
It moves AI from personal productivity toward business productivity.
So What Do Humans Still Do?
Quite a lot.
Both conversations eventually came back to the same boundary.
AI is increasingly good at factual, repetitive, digital, administrative, and analytical work.
Humans are still essential when the work requires judgment, nuance, accountability, motivation, conflict resolution, coaching, trust, relationships, change leadership, and understanding the politics in the room.
AI may identify the unknown unknown.
A human still has to decide whether it matters.
AI may tell us a project is slipping.
A human still needs to have the conversation.
AI may surface the best available information.
A human is still accountable for the decision.
And that may be one of the most important implications for the future of project execution:
As AI becomes better at doing the mechanics of project management, humans need to become better at the human work of project leadership.
Over the next several posts, I'll dig into some of the specific ideas these leaders surfaced: where the low-hanging fruit is today, what AI should handle versus humans, the risks of moving too quickly—or too slowly—and why investment in AI needs to happen alongside investment in human capability.
Because this technology is changing incredibly fast.
The question is no longer whether AI will change how we execute projects.
The more useful question is:
What should we stop asking humans to do—and what do we need humans to become exceptionally good at doing instead?
