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 one declares, “We must use artificial intelligence,” he is not, in truth, issuing a decision so much as revealing an unformed question; for beneath this resonant phrase lies a web of deferred inquiries yet to be properly articulated. If recast with analytical precision, it unfolds first into a search for purpose before means: what is the underlying problem that compels such urgency, and what defect in reality do we seek to amend? Then it turns to place: where, within the fabric of work or thought, does a locus exist that truly warrants delegation to a system that imitates or augments cognition? It proceeds further to weigh merit: do the anticipated gains—speed, accuracy, and efficiency—outweigh the latent risks of dependency, error, or distortion in judgment? And finally, it confronts method: do we possess the understanding, the infrastructure, and the data required to employ such a tool with discernment rather than mere fascination?
At this juncture, it becomes evident that the statement is less a conclusion than a haste toward one; it reflects a desire to keep pace with the age more than a comprehension of its direction. Thus, to properly unpack it is not to amplify its enthusiasm, but to return it to its origin: to transform it from a vague proclamation into a disciplined inquiry that defines the need, locates the application, evaluates the consequences, and verifies the capability. For it is the well-formed question—not the hurried answer—that distinguishes those who wield technology from those who are unwittingly shaped by it. 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.
En el desarrollo de este ejercicio, la utilización de herramientas de Inteligencia Artificial permitió evidenciar tanto su potencial como sus limitaciones en entornos corporativos reales. Si bien la IA facilita la estructuración, análisis y consolidación de información, su efectividad depende directamente del nivel de estandarización, calidad y organización de las fuentes de datos. En este caso particular, la heterogeneidad en nombres de archivos, estructuras y formatos redujo la confiabilidad de los resultados automatizados, lo que hizo necesario complementar el proceso con validación y estructuración manual. A partir de esta experiencia, se identifican dos aprendizajes clave:
La IA no sustituye el criterio analítico ni la toma de decisiones; actúa como un habilitador que requiere direccionamiento estratégico.
Su uso efectivo exige previamente la estandarización de la información (especialmente en archivos como Excel), así como la construcción de prompts estructurados y orientados a objetivos específicos.
En términos prácticos, el valor de la IA se maximiza cuando se integra dentro de un proceso organizado, donde se define claramente:
Cuándo utilizarla (procesos repetitivos, estructurados o de alto volumen)
Cuándo no utilizarla (información no estandarizada o de alta criticidad sin validación)
Esta experiencia permitió no solo avanzar en la consolidación de la información, sino también establecer una base metodológica para el uso responsable y eficiente de la IA en procesos de gestión documental y análisis de datos. En conclusión, la IA no reemplaza el proceso; lo potencia cuando existe estructura. Su adopción efectiva en la organización requiere madurez en los datos, estandarización y un enfoque estratégico en su implementación. Ing. Juan Sebastian Velasquez Montejo Colombia (+57)3203840207 Saving Changes...
Currently there is a lot of pressure eveywhere to incorporate AI - as it has been a box to tick. Meaning that whatever we do it should include AI component. I don't say AI is not necessary - but I believe that what is important is to stop and ask a question always - if there areas that AI could help us our KPIs. And if someone says "we should use AI" , i wouldn't kill this, but i would like to understand if someone really believes in value AI can bring or it is just a box to tick. 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.
We just started to discuss the use of AI by exploring the potential as well as the regulatory implications in our line of business Saving Changes...
Anonymous
Great analysis, which can help problem solving in impactful ways Saving Changes...
Gonzalo ValentiPM Specialist| MONOPOL LTDA.La Paz, La Paz, Bolivia (Plurinational State of)
Recently I propose to my team member analyze IA as a selling solution by the use of agents. The answer was: we don't have time to experiment with something that is not proven yet and it is better to spend time in look up for conventional strategies. I realized that there is a fear of using technology or maybe we don't understand yet, how to use it. Saving Changes...
"we should use AI” is a great ask but in what sense or how we could leverage AI , what are the limitations of using AI when dealaing with the confidential data / projects. Because once you start feeding your project data , information, metrics, stats to AI then the data is no longer in your control and can be viewed by your competitors or taken advantage of. So before diving into using AI agents , there should be clearly established procedures around using AI with clear instructions of not to share the confidential data to AI.
The AI tends to work differently when the data from same scenario is fed into multiple interpretations, the output from the AI could be varied and cant be relied upon. So it all boils down, how you tailor your input with the precision to expect any plans, suggestions from the AI agents. It is always Human in the Loop and Human in control. The other aspect to watch out when conversation PM things with AI is the use of words, since the AI closely depend on the sentiment analysis and based on that the AI agents could give you suggestions which are totally irrelevant to your projects. Saving Changes...
PMI Team Member
