If AI Can Do More of Project Management, Why Do We Still Need a Project Manager?

Exactly, Syed.
I would only be cautious about grounding the Project Manager's distinctive contribution in what AI cannot currently do, because those boundaries will continue to move.

AI may become increasingly capable of comparing trade-offs, modelling stakeholder preferences and recommending courses of action. But that still leaves a deeper organizational problem.

A supplier delay is rarely only a scheduling problem.
It may affect contractual commitments, customer expectations, cost, risk exposure, technical dependencies and relationships across several organizational boundaries.
No single analysis necessarily determines how those consequences should be reconciled.

That is where I see the Project Manager's contribution becoming more distinctive: not simply making the decision that AI cannot make, but helping the project system integrate distributed knowledge, surface competing consequences, preserve coherence across boundaries and ensure that decisions remain connected to the outcomes the project exists to create.

So the enduring distinction may be less human judgment versus AI analysis, and more the organizational capacity to turn distributed intelligence into coherent action.

I agree, Prasad.
And I think that distinction is fundamental.

As AI automates more of the administrative and analytical work associated with project management, it may actually make it easier to see what the role contributes beyond those activities.

I would take your point about raising the bar one step further.
The question is not only what better project leadership looks like, but what organizational function still needs to be performed when planning, monitoring, reporting and analysis require progressively less human effort.

Projects still have to translate strategic intent into coordinated action, integrate knowledge distributed across organizational boundaries, navigate interdependencies and competing priorities, sense when assumptions or conditions have changed, and preserve coherence as decisions become increasingly distributed between people and AI.

That suggests a deeper shift in the role: from managing the machinery of project execution toward helping the temporary organization remain coherent and adaptively viable.

In that sense, AI may not simply raise the bar for Project Managers.
It may help reveal more clearly what is genuinely distinctive about their organizational contribution.

Thank you, Stelian.
I agree with your distinction between AI and automation. AI is not synonymous with automation.
What matters for the question I am exploring is that AI can automate, augment or redistribute activities previously performed by Project Managers.

I would go one step further on the human side. Rather than grounding the future relevance of the Project Manager primarily in capabilities that remain exclusively human, I think the more durable question is whether an organizational function continues to be necessary as the boundary between human and AI capabilities evolves.

Projects still need strategic intent to become coordinated action, interdependencies and consequences to be connected across boundaries, competing priorities to be reconciled, changing conditions to be sensed, and distributed decisions to remain sufficiently coherent around the outcomes the project exists to create.

That integrative function does not disappear simply because some of the capabilities supporting it can be performed by AI.
In fact, how it is performed may increasingly involve both humans and AI.

So perhaps the distinctive contribution of the Project Manager should not be defined by what AI cannot do, but by the organizational problem that still needs to be solved.

And thank you for bringing CPMAI into the discussion.
It adds another useful perspective on how AI is changing the context in which project professionals operate.

Thank you, Emma.
I agree that uncertainty, competing priorities, stakeholder relationships, trust and accountability take us much closer to the real question than the administrative activities increasingly supported by AI.

I would add one distinction, though. I am cautious about defining the Project Manager's future relevance primarily by what AI cannot do. Those capability boundaries may change, and capability itself is different from authority or accountability.

An AI system may become increasingly capable of analysing trade-offs, supporting negotiation or recommending decisions without thereby acquiring legitimate authority to make a particular commitment or becoming accountable for its consequences.
At the same time, that authority and accountability do not necessarily reside with the Project Manager alone. In adaptive organizations, they may be distributed across sponsors, teams, governance bodies and other actors.

That is why I think the deeper question is organizational rather than simply human versus AI: who or what connects distributed knowledge, authority and decisions across boundaries so that the project remains coherent as conditions change?

The Project Manager's distinctive contribution may increasingly lie in helping perform that integrative function: translating strategic intent into coherent action while keeping consequences, responsibilities and outcomes visible, rather than personally owning every decision.

AI can change how that function is performed without making the function itself disappear.

Absolutely agree. AI may automate the mechanics of project management, but it cannot replace the leadership behind it. The real value of a project manager lies in navigating ambiguity, influencing stakeholders, resolving conflicts, making ethical trade-offs, and taking accountability when the path forward isn’t obvious.
The future isn’t AI vs. Project Manager—it’s AI-augmented Project Management, where AI handles the data and patterns while humans provide judgment, empathy, leadership, and accountability. The best project managers won’t compete with AI; they’ll use it to become better decision-makers and stronger leaders.

I agree, Carmia.
Responsible AI oversight adds an important dimension to the evolving role.

I would extend your point beyond validating AI outputs.
An output can be technically accurate and still be inappropriate for a particular decision because the data, assumptions, context, objective or decision criteria are incomplete or misaligned.
The governance question is therefore not only whether the answer is correct, but whether AI is being used appropriately, with sufficient evidence, transparency and meaningful capacity for human intervention before consequences become difficult to reverse.

I also think your point about documenting how recommendations influenced decisions is especially important.
As AI becomes more embedded in project work, organizations need to preserve traceability between evidence, AI-generated analysis, human judgment, decisions and resulting commitments.
Otherwise, accountability can become increasingly difficult to establish precisely as decision preparation becomes more automated.

I would only avoid placing that responsibility on the Project Manager alone.
Depending on the decision and its materiality, responsible oversight may need to be distributed across project leadership, sponsors, governance bodies, technical specialists and other accountable actors.

So perhaps the Project Manager's distinctive contribution is not to become the final verifier of AI, but to help ensure that AI-enabled work remains governable: that assumptions can be challenged, material influence can be traced, decision rights remain clear, and meaningful human intervention remains possible when it matters.

Your final point captures the risk well: automation can accelerate good work, but without adequate governance it can also accelerate flawed reasoning and its consequences.

I agree, Mayte.
This is very close to how I see the shift.

Better schedule intelligence, dependency analysis and scenario modelling are valuable precisely because they can release attention from work that does not necessarily require the Project Manager to be its primary producer.
The interesting question is where that attention should then move.

The areas you identify, strategy-to-execution translation, decision rights, trade-offs, organizational sensing and stakeholder alignment across boundaries, may look like separate higher-value activities, but I think they point to a deeper common function: integration.

Projects increasingly operate across distributed knowledge, authority, priorities and capabilities.
No single actor needs to own all of those elements, but the project system still needs them to remain sufficiently connected for coherent action to be possible as conditions change.

That is where I see the Project Manager's contribution becoming more distinctive: not as the primary processor of project information, or necessarily the owner of every decision, but in helping perform the integrative function through which the temporary organization can sense, decide, coordinate and adapt around the outcomes it exists to create.

For me, the important point is that the future relevance of the role does not need to depend on judgment, influence, context or organizational navigation remaining exclusively human.
The more durable basis may be the integrative organizational function itself, however the division of work between humans and AI evolves.