Yes. This is what happened to us exactly when we heard from our chairman asking us to work on the AI migration project without any further input. We brainstormed internally but no clue at the time. But as the time pass by, several AI tools started coming and we got a good idea now where to use them. Now, we are trying to use wherever possible from extracting the meeting, extracting invoices, image detection, coding, documentation, policy making etc. It is part of our everyday work now. Saving Changes...
What’s really happening most of the time is: AI is being seen as a buzzword or a magic wand, and people want to align with it. But if you peel the onion a bit, the conversation quickly shifts from AI to "we have a problem we’re trying to solve”
So, unpack it in three layers: 1. Reframe from “AI” to “Problem” - start by asking: What specific outcome are we trying to improve? What’s broken today?
Because often: They’re trying to map a problem to AI, instead of first defining the problem clearly.
2. Classify the Problem Correctly - once the problem is clear, categorize it: for example,
Analytics problem (dashboards, insights, reporting) Automation problem (workflow, repetitive tasks) True AI problem (prediction, language, reasoning, decision support)
In many cases: What’s being asked as “AI” is actually analytics or automation in disguise
3. Align the Right Solution (Not Always AI)
Only after that do bring AI into the picture—if it truly adds value.
This approach keeps the conversation grounded and ensures we’re solving the right problem, in an appropriate way. Saving Changes...
The thing is that AI is only a tool. First, you need a clear undertanding of the problem. that's how you got clarity of when use AI, it would be needed as far as it help you to solve a business problem
When this question is asked, the first thought that comes to my mind is to use LLMs such as OpenAI—for tasks ranging from correcting grammar to interpreting data in various ways. However, I am not sure how to implement this effectively or which specific AI applications should be used to address our project management requirements. In particular, I am looking for AI tools that can help manage workflows efficiently without spending excessive time analyzing manual reports Saving Changes...
AI can help in landscape projects in area like Water demand prediction, smart irrigation, scheduling and automated CAD/ BIM design integration . if we integrate it with tool like AUTOCAD or Land F/X, it improves efficiency significantly.
Saving Changes...
Rita AveritteSr Project Manager| HCALittle Elm, Tx, United States
Mar 25, 2026 1:39 PM
Replying to Ivan Samuel Santana
...
I participated in one of these projects, which involved developing a chatbot to help respond to citizens’ inquiries. Shortly afterward, a manager suggested introducing AI as a tool so that the chatbot could respond intelligently. However, nothing was said about the need to update and reorganize all the information available on the website so that the chatbot could provide coherent and well-structured answers. Before considering the application of AI, it is essential to first assess whether we have the appropriate environment for the project to be successful.
Thats definitely true. Environment is everything for success Saving Changes...
Wayne HoughtonStrategic Programme Leader| Retail & Financial Services IndustryCape Town, Western Cape, South Africa
These days almost everything is being referred to as AI, but might in fact be much simpler automation technology. However, therein lies the challenge, in understanding if it is genuinely an AI problem you are solving, and what type of AI solution you need to reference.
In my limited experience with AI, I have found that in my organisation it has largely been quite bespoke and insulated in specific areas, but recently, in the last month signalled a much broader enterprise-wide signal that we will see a much deeper and wider adoption of AI into the fabric of our business. One of our HR Execs was appointed as Chief AI Officer - and this signalled to me a larger scale intent highlighting the change and AI adoption at every role in the organisation, and not just specific areas. Saving Changes...
MANUELA BONIZZIGlobal Medical Information Planning Lead| TAKEDACham, Zg, Switzerland
When someone says, “We should use AI,” I try to unpack what they actually mean by first clarifying the intended outcome. A useful way to separate “AI” conversations is by the job it’s expected to do—automation, prediction/insight, or content generation. If the ask is about insights and forecasting, we’re usually talking about predictive AI/ML. If it’s about creating text, images, or code, it’s likely generative AI. And if it’s about reducing manual work, it may be workflow automation (possibly with AI embedded). Where teams get stuck is treating all of this as simply “AI” without a concrete use case. That ambiguity tends to create unrealistic expectations, the wrong tool choices, and misalignment with business goals. I’ve been in plenty of discussions where different stakeholders meant different things by “AI”—you can usually spot it when requirements stay vague (e.g., “use AI to improve efficiency”). In those moments, I shift the conversation by asking a few clarifying questions: What decision or task are we improving? What does “better” mean (time, cost, quality, risk)? What data do we have (and is it usable)? What output do we need (a prediction, a draft, a recommendation, an automated step)? Ultimately, the goal is to move from “using AI” to solving a specific business problem with the right approach—and being explicit about the expected outcome. Saving Changes...
We understand that AI can be helpful, but we need to be clear about which specific AI system will be used, since there are many available. Saving Changes...
One of the biggest signals for distinguishing different types of AI work is the expected outcome—whether the goal is automation, prediction, or content generation. For example, if the focus is on insights and forecasting, it’s likely predictive AI; if it’s about creating text, images, or code, it points to generative AI. What often goes wrong is when everything gets labeled simply as “AI” without clarifying the use case. This can lead to unrealistic expectations, poor tool selection, and misalignment with business objectives. I’ve definitely been in conversations where “AI” meant different things to different stakeholders. Usually, I notice it when requirements are vague—like “we should use AI to improve efficiency” without defining how. That’s when I step in to ask clarifying questions about the problem we’re trying to solve, the data available, and the desired outcomes. In my experience, the key is to shift the conversation from “using AI” to “solving a specific business problem with the right AI approach.”
I agree, the focus needs to be on the problem, not the solution. This is a problem we've been having in projects since ... forever! We have a new, powerful tool, AI, and everyone wants to jump on the bandwagon because it's money saved, right? But the focus needs to be on the problem, and how AI will solve it. Saving Changes...
PMI Team Member
