Pham Van Phuong
Highly relevant topic.
The integration between AI tools and human judgment is one of the biggest challenges (and also a strategic opportunity) in modern project management.
In my experience, I see AI as a valuable assistant, capable of supporting pattern recognition, scenario simulation, and risk anticipation.
But human judgment remains irreplaceable, especially in zones of ambiguity, ethical impact, or political project dynamics.
Three practical principles I apply:
- Human at the center, AI as support
AI doesn’t replace decision ownership.
It offers complementary insights.
Preserving that balance is essential to maintain decision legitimacy and team trust.
- Transparency and explainability
I only integrate AI recommendations that I can explain to the team, to the sponsor, or to myself. If it’s not understandable, it’s not reliable.
- Clear governance
We define explicit criteria for when to trust AI, when to validate through experience, and when to slow down and consider ethical, cultural, or relational dimensions.
I’ve seen situations where AI led to poor decisions, for instance, predictions based on historical data that failed to reflect recent contextual or organizational changes.
To prevent that, we’ve adopted a lightweight internal framework with 3 key questions before accepting any AI-driven recommendation:
- What might the AI not be seeing?
- Who will be affected by this decision?
- Will this decision strengthen or weaken team trust?
When used with critical awareness, AI can expand our field of vision.
But without human criteria, it may reduce decisions to algorithms that don’t know the terrain.
Thank you for raising this topic
I’m following the conversation with great interest.
