To me, the biggest signal is the cognitive verb being used. If a stakeholder says they want AI to "recognize" or "flag" something, we’re talking about Pattern Recognition. If they want it to "predict" or "forecast," we’re in Predictive Analytics.
I knew I was in trouble once when a client asked for an "AI chatbot" (Conversational Pattern), but then expected it to perfectly forecast their quarterly revenue (Predictive Pattern). It’s a classic "lumping" mistake. One is built to be a great conversationalist; the other is built to be a math genius. When you lump them together, you end up trying to use a hammer to turn a screw.
What I do now: Whenever someone says, “We should use AI,” I immediately ask: "Are we trying to offload a repetitive decision, or are we trying to generate new content from scratch?" Getting them to define the task instead of the tech usually tips me off to whether they’ve been sold a "magic box" by a vendor or if they actually have a specific bottleneck we can solve. Without that distinction, you’re just setting yourself up for a project where the data doesn't match the goal. Saving Changes...
Artificial intelligence that cannot possess emotional intelligence will always remain dependent on humankind Saving Changes...
Pouviraj ShambooFunctional Manager| SBM Bank MauritiusLower Dagotiere, MO, Mauritius
Mar 25, 2026 4:50 AM
Replying to Douglas Boyd
...
It is recognised that AI can assist, but we need to obtain clarity as to what AI system is to be used as there are many.
Indeed, value added and risks minimization are great benefits for an organization. However, we will need to define the problem well. Saving Changes...
Aamer AbbasAamer Abbas, PMP| National Australia BankMelbourne, Vic, Australia
Feb 19, 2026 1:05 PM
Replying to Luis Branco
...
Great question.
When someone says “we should use AI,” the conversation is rarely about technology itself. It is usually about pressure for speed, efficiency, innovation, or competitive leverage. The first step is to clarify intent.
Three signals help distinguish what is really being asked.
First, decision proximity. Is AI automating a task, augmenting human judgment, or moving toward managing objectives autonomously? These are fundamentally different categories of work. The closer AI gets to consequential decisions, the stronger the need for governance, traceability, and explicit oversight.
Second, problem clarity. Is there a clearly defined business problem with measurable impact, or is AI being treated as the starting point? When the solution precedes the problem, misalignment and inflated expectations follow.
Third, accountability design. Who owns the outcome if an AI-driven recommendation fails? When responsibility becomes diffuse, risk scales faster than performance.
In many organizations, “AI” simultaneously means efficiency, experimentation, and cost reduction to different stakeholders. Misalignment becomes visible when decision flows and ownership are unclear. A common tipping point is when stakeholders use the same word “AI” but describe different success metrics.
The real shift is not from manual to automated. It is from “man in the loop” to “man in control.” Without deliberate design of responsibility, capability increases while accountability erodes.
Clarity of purpose, category of AI work, and ownership separates disciplined transformation from technological noise.
Can't agree more. However, in my experience, most of the time when business says "Let's use AI", they usually mean quick turnaround. Saving Changes...
A common mistake is when everything is broadly called “AI” without defining the actual use case. This often creates unrealistic expectations, leads to choosing the wrong tools, and causes misalignment with business goals. I’ve personally seen situations where different stakeholders had completely different meanings of “AI.” It usually becomes obvious when requirements are too vague, such as saying, “we should use AI to improve efficiency,” without clearly explaining what problem needs to be solved or what outcome is expected. Saving Changes...
Cammie ParkerProgram Manager| RaytheonVail, AZ, United States
Perhaps the question being asked isn't truly about the solution so much as it is about an improved definition of the problem. A huge fan of Chat GPT, I have found that the earlier I can provide the things I am uncertain of or don't fully understand to the AI client, the sooner I gain a fuller understanding of what I don't know at the outset. What I think I know (being the smart human in the conversation) is often wrong, or my perspective is slewed by my limited human knowledge.
In allowing AI to help me better define the problem, I am able to assess what I have the capability and resources to address or resolve, then subsequently map a path toward a minimum viable product, solution, or even a set of solutions depending on what I am trying to resolve.
I don't ask AI to fix it, I ask AI to help me understand the problem, then use phased approaches (using AI) to find workable solutions. Saving Changes...
Anonymous
It is known that AI can help, but we need to obtain clarity as to what and when AI system is to be used Saving Changes...
Danna DanfordProject Manager | Operations & Client Success SpecialistIRVING, United States
Because "AI" is currently such a massive umbrella term, I frequently see it used as a placeholder for very different concepts depending on who’s talking.
It usually takes a few "Wait, what do you mean by that?" questions to get everyone on the same page. Without those, you end up with one person building a rocket ship while the other just wanted a better toaster.
Examples:
Automation vs Intelligence A teammate says, "We need to add AI to this spreadsheet," but they actually just mean a complex If/Then macro. The disconnect, for many, AI is synonymous with automation—anything that saves them from clicking a button. For others, it specifically means Machine Learning, where the system improves itself without being explicitly programmed for every scenario.
Product vs Research In a professional setting, a Project Manager might be looking for a finished tool to increase ROI, while a Developer is talking about latent space and model weights. The disconnect, one sees a utility; the other sees a math problem. Saving Changes...
When it comes to AI or any other technology, it should solve a problem or enhance what we have, and of course you need to understand the problem then you can find a solution for it Saving Changes...
Alan RuskinProgram Director| McMurdoArnold, Md, United States
When something doesn't 'look' like something the creator would produce it's a clue that AI was used. I think it's important that we don't let AI do all the work and take our personality and creativity out of our work.
When someone suggests using AI, they seems to be looking for a shortcut. Sometimes it's meant to have a tool create everything, other times it's meant to have tool check work that was already done. Saving Changes...
PMI Team Member
