Yes, it happens sometimes. Someone means a chatbot, another means image generation, another means recommendation engines. My signals: what capability is being discussed, eg. conversation vs. pattern-matching vs. physical action, what failure mode worries them (hallucination vs. bias vs. safety), and whether a human stays in the loop. That's why you have to create an "AI baseline" of understanding across the team. Saving Changes...
EILEEN WANGPurchasing Manager| NoneBrisbane, Australia
What is behind using AI is a purpose or a need that an organisation or an individual going to make changes..
Saving Changes...
Anonymous
People are not familiar with what AI means so figuring out the problem or workflow to improve or change defines the mental model to then determine if and where AI can be used within the project. Saving Changes...
Anonymous
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.
When someone suggests we use AI to solve a problem, I try to first understand the root cause of the issue to determine whether AI is the right answer. Through a series of probing questions, I assess where the deficiencies are originating and then consider how AI may be leveraged to meaningfully address the issue. Saving Changes...
When someone says, “we should use AI,” I treat it as a signal rather than a solution. It often reflects a desire for improvement, whether that means speed, efficiency, or innovation, but the technology itself is secondary.
The first step is to clarify the problem we are trying to solve and the outcome we want to achieve. Without that clarity, AI risks becoming a buzzword instead of a tool. I also consider feasibility: do we have the right data, processes, and accountability in place to make AI effective?
From there, I translate the broad idea of “AI” into practical options such as machine learning for predictions, natural language processing for text, computer vision for images, or automation for repetitive tasks. Sometimes the right answer is not AI at all.
My usual response is to define the problem first, then decide if AI is the right fit or if another approach will deliver more value. This reframes the conversation from hype to business value and keeps stakeholders aligned on intent, feasibility, and responsibility. Saving Changes...
When someone drops the phrase, “we should use AI,” it’s rarely a straightforward technical requirement. More often, it’s a placeholder for a mix of excitement, pressure, or a specific underlying problem. Saving Changes...
Anonymous
When someone says, “We should use AI,” I try to understand the underlying need rather than focusing on the technology itself. Usually, they're expressing a desire to solve a problem, improve efficiency, reduce costs, gain insights, or stay competitive. The first step is to ask questions like: What problem are we trying to solve? and What outcome are we hoping to achieve? Once the goal is clear, we can determine whether AI is actually the right solution and how it can add value. Saving Changes...
Anonymous
Signals usually come from the type of problem, the kind of output expected, and the level of reliability required. If the work is about predicting an outcome from historical patterns, that points to predictive analytics or machine learning. If it is about generating text, images, summaries, or drafts, that suggests generative AI. If it is about automating repeatable rules-based steps, that may be workflow automation rather than AI in any meaningful sense. The confusion starts when those categories are treated as interchangeable. What tends to go wrong is that teams choose tools before defining the job to be done. “Use AI” can mean automate a process, surface insights from data, support decisions, draft content, or create a chatbot, and each has different data needs, risks, governance requirements, and success measures. A common signal that people mean different things is when one group talks about efficiency, another talks about innovation, and a third expects human-level judgment. That mismatch usually shows up in vague requirements, unrealistic expectations, and weak evaluation criteria. A practical way to sort it out is to ask: what task are we trying to improve, what input does the system need, what output would be useful, and how much error is acceptable? Those questions separate experimentation from production use and help reveal whether the need is for automation, prediction, generation, decision support, or some combination. Saving Changes...
Anonymous
I unpack it by translating “use AI” into a concrete business need. The first distinction is whether the team wants to generate, predict, classify, recommend, or automate. Those are very different kinds of work, and each implies different data, tools, risks, and success measures. I then look for signals in four places: the problem being described, the input available, the output expected, and the tolerance for error. If the ask is “draft responses” or “summarize documents,” that points toward generative AI. If it is “forecast delays” or “identify churn risk,” that is closer to predictive analytics. If it is “route requests” or “trigger actions,” that may be workflow automation with or without AI. What usually reveals the real ask is a simple reframing: what decision or task are we trying to improve, and what would a useful result look like? Once that is clear, it becomes easier to tell whether the team actually needs AI, what type of AI fits, and how to evaluate whether it adds value. Saving Changes...
I unpack it by translating “use AI” into a concrete business need. The first distinction is whether the team wants to generate, predict, classify, recommend, or automate. Those are very different kinds of work, and each implies different data, tools, risks, and success measures. I then look for signals in four places: the problem being described, the input available, the output expected, and the tolerance for error. If the ask is “draft responses” or “summarize documents,” that points toward generative AI. If it is “forecast delays” or “identify churn risk,” that is closer to predictive analytics. If it is “route requests” or “trigger actions,” that may be workflow automation with or without AI. What usually reveals the real ask is a simple reframing: what decision or task are we trying to improve, and what would a useful result look like? Once that is clear, it becomes easier to tell whether the team actually needs AI, what type of AI fits, and how to evaluate whether it adds value. Saving Changes...
PMI Team Member
