What I have experienced within the industry is that there is no clear understanding in the purpose and/or benefits carried through AI; therefore, becomes more a trend to "have" rather than a specific solution with a calculated ROI. In that context, companies get confused and if not aligned in the purpose and expected solution with expected performance and ROI, quicky gets discouraged and disband initiatives beyond LLM platforms or transcript assistants Saving Changes...
Lusizo MsibiLearning and Development Coordinator| Training ForceEkurhuleni, South Africa
The term “AI” has become so broad that it often creates more confusion than clarity in project environments. One of the key signals I use to distinguish different types of AI work is the expected outcome, is the goal automation, prediction, insight generation, or decision support? Each of these requires very different approaches, tools, and levels of data maturity. Another important signal is the data requirement. If a solution depends heavily on structured historical data, we’re likely about machine learning. If the focus is on language, communication, or content generation, then it leans more toward generative AI. Misalignment often happens when stakeholders assume all AI behaves the same way, regardless of these differences. I’ve definitely been in conversations where “we should use AI” meant completely different things to different people. The biggest red flag is when there’s no clearly defined problem, only a desire to apply AI because it’s trending. That’s usually when expectations become unrealistic, timelines slip, and value isn’t delivered. What helps is reframing the conversation:
What problem are we solving?
What decision are we trying to improve?
What data do we actually have?
Only then does it become clear whether AI is even the right solution, and if so, what kind. AI is powerful, but without clarity, it quickly becomes a buzzword rather than a tool for real impact.
As a Project Manager and Data Scientist working within investment management and audit functions, I’ve found that the biggest challenge—and opportunity—lies in defining what 'AI' actually means for our specific workflows. In my day-to-day work, I treat AI as a two-sided tool:
Predictive AI: I use Python-based models to develop systems like 'Audit Risk Scoring.' Here, AI helps prioritize which projects or transactions need immediate attention based on historical risk patterns.
Generative AI: I use LLMs to streamline the 'administrative heavy lifting,' such as drafting official notes for audit objections or refining technical code for Streamlit applications.
The key takeaway from my experience is that AI doesn't replace the Project Manager; it upgrades our role to 'Human-in-the-Loop' validators. Especially in financial administration, we cannot rely on a 'black box.' We must validate the AI's output against regulatory frameworks to ensure accountability and precision. I’d love to hear how others are balancing technical AI implementation with the rigid compliance requirements of the financial sector! Saving Changes...
On a recent project, we hired an "AI vendor" to help process some critical workflows. An unspoken expectation of half the team was that the AI functionality would allow the workflows to be dynamic and more adaptable over time, while the other half believed AI would simply provide greater accuracy to existing processes than was previously possible. This mismatched expectation led to a lot of refactoring of the project over time, delays, and time wasted as individuals were working toward different understandings of the goal Saving Changes...
If someone says, we should use AI. The question is, what are we trying to accomplish? What value can AI add. We still need to understand the need in order to validate the data. Saving Changes...
I agree that for many organizations, AI means efficiency and cost reduction, but it's not just that; fundamentally, it's about making every activity better and better using the tools we need—in this case, AI—and we as project managers must learn and adapt quickly to generate value in a more comprehensive way. 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.
This is a great response to a project ask - with great granular detail! I love these requests for more tools, but have tended to ask more open ended questions of "why this tool in particular" or "which problems & issues do you think it will help with"? Saving Changes...
George MathewPM Specialist| GgRooty Hill, Nsw, 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.
The biggest question should be on client or user requirement. AI may help solve it or may not help. That is upto the design and technical team to decide based on the requirement. Saving Changes...
PMI Team Member
