By Nicole Petite
November 11, 2026
Are We Moving Too Fast with AI Without Setting Boundaries?
Artificial intelligence is changing the way we work, make decisions, manage projects, and conduct business. Organizations are investing in AI because they want to become more efficient, reduce costs, improve performance, and stay competitive. And I understand why. The possibilities are tremendous.
But there is something about this rapid movement toward AI that concerns me.
Are we taking the time to establish who is responsible for what AI is allowed to do before we put it to work?
We hear a great deal about Human-in-the-Loop (HITL), where a person reviews, corrects, approves, or intervenes in the work AI produces. I absolutely believe HITL is necessary. But my concern is what happens before we ever get to that point.
Who established the rules? Who determined what was acceptable? Who examined the potential biases? And, most importantly, was that person qualified to make those decisions?
These are the questions that led me to develop what I call Human-in-Place (HIP).
What Is Human-in-Place (HIP)?
Human-in-Place is my proposed methodology for ensuring that qualified human oversight is established before an AI system is developed, configured, or placed into operation. It means having the right person, or team of qualified individuals, responsible for defining what the AI system is supposed to accomplish, how it should perform, what it should never be allowed to do, and what safeguards must be in place to protect the people and organizations affected by its decisions.
And let me be clear. I am not suggesting that just anyone should fill this role. The person responsible must understand quality, ethics, bias, risk, and the intended business or operational outcome. That person must also have the critical-thinking skills and authority to challenge assumptions, question recommendations, and determine whether AI is even the right solution in the first place.
HIP establishes what AI is expected and permitted to do. HITL provides human involvement as AI performs.
I believe we need both.
Why Should This Matter to You?
Because you could find yourself on either side of the consequences.
Imagine applying for a job and being denied an opportunity because an AI system used historical information that reflected unfair hiring practices. Or imagine applying for a loan and being rejected because the system evaluated information using standards that were never properly examined for bias.
Now, put yourself on the other side.
What if you are the organization responsible for implementing that system? What if a decision made using AI results in discrimination claims, legal consequences, financial losses, or damage to your organization’s reputation? And what if the issue could have been identified before the system was ever put into operation?
That is exactly why I believe HIP deserves attention. An AI system can function exactly as it was programmed and still produce the wrong outcome because the original requirements were incomplete, inappropriate, or poorly defined.
You cannot expect a person to properly evaluate an AI recommendation if no one established the right standards for evaluating it in the first place.
How Would HIP Work?
My methodology focuses on six responsibilities that should be established before AI implementation and maintained throughout its lifecycle.
- Establish the purpose. What problem are we trying to solve? What results do we expect? And do we actually need AI to solve it?
- Define expected performance. What should the system do? What information can it use? What results are acceptable, and what limitations must be established?
- Address ethics and bias. Who could potentially be harmed? Are we using information that could lead to unfair treatment? What safeguards need to be in place?
- Establish accountability. Who has the authority to approve decisions, question results, stop the system, or escalate concerns?
- Determine how success will be validated. How will we test the system? What evidence will demonstrate that it meets our standards? And what happens if it doesn’t?
- Maintain oversight. Once the system is operating, who ensures it continues to meet those expectations as information, technology, and organizational needs change?
That last responsibility is especially important to me. HIP should not be something we document at the beginning of a project and then put away. Human accountability must remain in place throughout the lifecycle.
What Does This Have to Do with Project Management?
Quite a bit!
As a project management professional, I have spent years working with requirements, scope, risks, quality standards, stakeholder expectations, and accountability. We would not intentionally begin a major project without determining what we are trying to accomplish, identifying the requirements, understanding the risks, and establishing what success looks like.
So why would we approach implementing artificial intelligence any differently?
If we fail to establish the right requirements at the beginning, we risk spending time and money developing a system that may never have been properly scoped to solve the problem. Now add the possibility that the system could influence someone’s employment, financial stability, access to services, or future opportunities.
I believe HIP can help organizations apply disciplined project management thinking to AI governance while ensuring that qualified individuals establish the appropriate ethical, operational, and performance expectations. My goal is to build on those principles by identifying a clear human responsibility that organizations can put into practice.
My Position: Put Human Accountability in Place First with a Qualified Person
I support artificial intelligence and the opportunities it creates. I teach AI, I advocate for innovation, and I believe this technology can help organizations accomplish some remarkable things.
But I also believe we have a responsibility to make sure we are using it correctly. And that responsibility should not begin after the system has already made a recommendation, influenced a decision, or negatively affected someone’s life. It should begin before the system is ever put to work.
Human-in-the-Loop is important because people need to remain involved in AI processes. Human-in-Place is my proposed approach to making sure qualified people have established the expectations, safeguards, and accountability that guide those processes from the beginning.
We should not wait for AI to make a mistake before deciding who was responsible for preventing it.
That is the conversation I want organizations, technology leaders, project managers, and policymakers to have. Because in my view, the future of responsible AI should not be determined only by what technology is capable of doing. It should also be determined by what qualified humans decide that technology should be allowed to do. And before AI takes its place, we need to make sure the right humans are already in place.
Nicole Petite is a certified PMP and PMI Authorized Training Partner (ATP) with 8 years of Agile experience. She has led teams using Scrum and Kanban to deliver projects efficiently and adapt to changing requirements. Nicole works with organizations to align Agile practices with business goals, improve team performance, and ensure project success. She is the CEO of ProjIT Solutions, CEO of Nicole Petite Professional Training, and VP of Professional Development for PMI North Alabama. She is passionate about equipping professionals with the tools, training, and mindset to lead with confidence in today’s dynamic project environments.
