It starts with asking further probing questions. Where the idea to use AI come from and what situation are they trying to solve or process to improve. The more you know of the idea the best way you identify what is needed, which tool is best or what connectors will help you deliver the intended project. Saving Changes...
I believe that digital transformation, emerging trends, and new technologies invite us to continuously look for new and better ways of doing things. When someone says, “we should use AI,” I see it as an invitation to look at the project from a systemic perspective and understand what we are really trying to achieve. The first step is to unpack the request by asking: What problem are we trying to solve? In which project phases could AI add value? Could it help us accelerate interactions, improve decision-making, automate repetitive tasks, analyze information, or deliver better and faster outcomes? However, using AI should not be a goal in itself. We need to evaluate whether it actually creates value, considering factors such as data quality, risks, costs, security, and the impact on the project team and stakeholders. Most importantly, AI does not replace human judgment. It requires human oversight and intervention to validate the quality, accuracy, and relevance of the information it produces. Saving Changes...
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.
Great insightful analysis Saving Changes...
Jaime CastilloProject Management| AI INVERSIONES PALO ALTO II SACLIMA, LIM, Peru
When we say we are going to use AI in a project, we must have clarity and consensus on what is expected from that scope. We can generally break this down into three different levels of application:
Support: Using AI simply to assist and facilitate the work of managers and operational teams.
Implementation: Leveraging AI to build the solutions or services that form part of the project’s deliverables.
Core: Integrating AI as the very essence of the solution itself.
This variety of possibilities opens up a crucial space for alignment, discussions, and negotiations among different stakeholders. What is clear, however, is that today we cannot move forward without considering AI in our projects. The sheer volume of data, the demand for precision, and a truly holistic approach are far better guaranteed when we include it. The real question is: Are we all truly prepared for this shift, or are we still just looking at AI as an upgraded chat tool? Saving Changes...
When someone says "we should use AI," the ask is almost never the ask. What's usually underneath: Who's saying it, and why: a staff capacity complaint, a board chasing FOMO after a peer org's announcement, or a funder narrative need are three different projects wearing the same sentence. What they mean by "AI". We get them naming an actual task ("what would be different a month from now for your org?") rather than the word itself, since it could mean anything from a chatbot to auto-generated newsletters. What problem it's actually solving. Saving Changes...
Haitham HelmyFunctional Manager| Augustus Mediamansoura, DK, Egypt
Does AI add value here? — Compare it with simpler solutions such as rules, search, automation, or better UX. AND What decision or task should AI improve? Saving Changes...
translating that enthusiasm into a clear business problem, data plan, and accountability framework. Saving Changes...
Sandra ChouinardSenior OCM SpecialistMartinsville, in, United States
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.
AI is the latest 'fad' word. Everyone throws it around without understanding what it is/can be, let alone how it is able to enhance or derail their situation depending on how it is used. Incorrect AI infulience can lead to so much misdirection and added work. Unravelling it can become a nightmare. Saving Changes...
MUHAMAD FUA AD BIN AZIZSenior Project Management| UEM SUNRISE BERHADSungai Petani, 2, Malaysia
When someone says, “We should use AI,” we should first understand the perspective from which that person is coming, what they actually need, what problem they are trying to solve & what outcome they expect.
Different people may use AI for different purposes & they may not necessarily expect the same output, level of detail, accuracy or approach as we do. Therefore, before using AI, it is important to understand the user’s perspective, objective, expectations & intended outcome.
AI is only as effective as the direction we provide. The right question is not simply “can we use AI?”, but rather “what do we need AI to achieve & what does a successful output look like for the person using it?” Saving Changes...
When someone says, “We should use AI,” the first step is to unpack the problem they are trying to solve and determine whether AI is suitable for that specific use case. Although AI capabilities often exceed those of traditional tools, they also introduce risks associated with a rapidly evolving technology. Therefore, AI adoption should not be approached as a one-size-fits-all solution. Mental models can guide decision-making, but organisations must also understand the basic differences and nuances among available AI solutions. For example, a third-party AI-enabled tool may perform the required task more effectively than an internally developed solution. In such cases, factors such as cost, implementation time, performance, security and control may determine the most appropriate option. The organisation must also assess its internal capabilities. This includes determining whether the team has the skills, infrastructure and capacity to implement, use, maintain and support AI tools embedded within its products or solutions. Adopting an advanced tool without the ability to manage it effectively may create operational difficulties and increase dependence on external providers. Another important consideration is vendor dependency. When a solution is built using models developed by frontier AI laboratories, the organisation becomes reliant on those providers for continued access, model performance, pricing, updates and availability. This reliance can become problematic when there are few alternative providers offering comparable or better capabilities. Therefore, the decision to use AI should also consider data ownership, intellectual property, interoperability, vendor lock-in and the organisation’s ability to migrate to another solution if necessary. Saving Changes...
PMI Team Member
