When someone says, “we should use AI,” how do you unpack what’s really being asked?
To avoid pursuing the 'magic' of AI without a clear foundation, I suggest we use the '5 Whys' to clarify the requirements. This will help us validate if AI is appropriate and ensure we implement a structured, effective solution.
When someone says "We should use AI", unpacking their thoughts and having a decent response is the first challenge. Many don't have any real understanding of AI, myself included in many aspects. The most recent suggestion I had was: "We need AI to plan and sort out manpower availability, to prevent gaps in services caused by workers absence and in many cases language barriers". My approach was to dig deep into my own mental model by exploring many years of experience to establish a start point, an end point and a plan for this specific problem, as well ask asking them to do the same. There is a lot of information to unpack, so as mentioned in earlier comments, assessing feasibility, technical, organizational, and governance constraints, before a scope is defined is a priority. We are still unpacking it but my takeaway on this is it's a cross team issue Saving Changes...
Whenever you hear someone say "Use AI," just ask them one simple question: "Are you trying to save time on writing reports, or are you trying to predict a problem before it happens?" That one question usually stops the confusion and helps everyone focus on what actually needs to be done. Saving Changes...
I think one of the biggest opportunities with AI is that it allows project managers to spend less time on repetitive administrative work and more time leading teams, solving problems, and engaging stakeholders. AI should be viewed as a tool that supports better decision-making—not as a replacement for critical thinking or human judgment. The key is learning where it adds value and where experience and leadership remain essential. Saving Changes...
The signals that help you tell different kinds of AI work apart depend first and foremost on the real need within the value chain. In the textile industry for example, "AI" doesn't mean the same thing depending on where you are in the business:
Is it a production chain management problem? → we're talking about autonomous systems and real-time monitoring
Is it a supply chain problem? → we're talking about demand forecasting and inventory optimization
Is it a quality control problem? → we're talking about pattern recognition and anomaly detection
The signal is always the nature of the business problem — not the word "AI" itself.
What tends to go wrong when everything gets lumped together is that you lose touch with the real, concrete need on the ground. The solution must adapt to the reality of the business — not the other way around. To avoid this, you need to bring in the operational team from the very beginning — they are the ones living the problem every day and who know exactly where things break down. The real balance to strike is building a solution that:
Answers the concrete need of the operational team on the ground
While giving the visibility that the executive is looking for to steer strategy and make informed decisions
When those two levels are aligned, the word "AI" finally has a precise meaning — because it is grounded in a shared reality that everyone understands. Saving Changes...
PMI Team Member
