top of page

AI Won’t Save Your Organization. Leadership Will.

Nov 4
6 min read

By Achille Ettorre



AI is being positioned as the answer to almost every business problem we face. It will improve productivity, reduce costs, transform customer experience, improve forecasting, automate routine work, analyze huge volumes of information, and surface patterns that would once have taken people days or weeks to uncover.


For finance professionals, the potential is enormous -- But there is a problem.


AI cannot fix unclear leadership. It cannot repair a broken culture. It cannot make poor data trustworthy and it cannot restore trust once it has been lost. In fact, AI does not simply inherit organizational weaknesses -- It can industrialize them.


That is why one of the most important questions leaders should be asking is not simply:

“What can AI do?” It is: “What happens when we use it?”


You Can’t Automate Accountability

The Air Canada chatbot case provides a useful example.


A customer asked the airline’s chatbot about its bereavement fare policy. The chatbot provided incorrect information, and the customer relied on it. When the matter was challenged, Air Canada argued that it should not be responsible for what its chatbot had communicated. The tribunal disagreed.


The chatbot was part of Air Canada’s website. It represented the company.


The lesson is simple:

You can automate an interaction, but you cannot automate away accountability.


That principle matters far beyond customer service. AI-generated information is increasingly influencing forecasts, pricing, hiring, customer communications, financial analysis, operating decisions and investment choices. When something goes wrong, an organization cannot simply say, “The algorithm made the decision.” Someone still owns the outcome.


For finance leaders in particular, this should feel familiar. We have always been accountable for assumptions, controls, models, inputs and decisions. AI does not remove that responsibility -- it raises the stakes.


AI Doesn’t Eliminate Judgment. It Moves It.


This may be one of the most important principles leaders need to understand about AI. Algorithms do not eliminate human judgment. They relocate it.


Consider the case involving iTutor Group.


According to the U.S. Equal Employment Opportunity Commission, the company’s application software automatically rejected female applicants aged 55 or older and male applicants aged 60 or older. The company later agreed to a $365,000 settlement. The system did what it had been configured to do and that was precisely the problem.

Someone made a judgment. That judgment was translated into software. The organization then allowed the system to apply that judgment repeatedly and at scale.


AI did not create the original decision -- It made the consequences faster, less visible and more scalable.


Finance professionals understand this dynamic better than most. We build models. We select assumptions. We determine which variables matter. We decide which risks are acceptable. We establish thresholds. We interpret outputs. AI does not eliminate those choices -- it can simply bury them deeper inside the system. That makes transparency and governance more important, not less.


The Business Case Isn’t the Whole Case


Finance professionals are trained to ask two fundamental questions:

  • What will this cost?

  • What return will we generate?


Those questions still matter, but AI requires several others.

  • Who carries the risk?

  • What happens to employees whose jobs or required skills change?

  • How is customer information being collected, analyzed and retained?

  • What new technology requirements are being imposed on suppliers and business partners?

  • What happens if an automated recommendation is wrong?


An AI initiative can clear every financial hurdle and still be a poor management decision.

That does not mean organizations should avoid AI, it means the business case needs to be broader than a spreadsheet. Governance, trust, adoption, security and organizational impact ultimately affect the economics of the investment as well.


Good governance is not a tax on innovation, it is what allows innovation to scale.


Shadow AI Is Already a Finance Issue


One of the most immediate AI risks inside organizations is also one of the least visible: shadow AI. Employees are rarely using unauthorized AI tools because they are trying to create risk. They are using them because the tools work. Someone discovers that an AI application can summarize a lengthy report in seconds. Another employee uses one to analyze transactions. Someone else uses AI to prepare a customer response, review a contract or build a financial model.


The risk emerges when confidential contracts, customer information, financial data, employee information or proprietary business knowledge are entered into tools the organization has never reviewed. Simply banning AI will not solve the problem. If the approved process is dramatically slower than the unapproved alternative, employees will find workarounds.


So the first governance question should be very practical:

What are our people already doing with AI?


Before an organization can govern AI effectively, it needs visibility into where AI is being used. You cannot govern a system you do not know exists.


A Practical Framework: The Five Cs

Ethical AI can quickly become an abstract conversation. Leaders need something more practical. Before approving, implementing or scaling an AI initiative, I recommend asking questions in 5 areas.


1. Consent

  • Do the people affected understand how their information is being used and what the system is doing?

  • Where appropriate, have they agreed to it?


2. Clarity

Can we explain what the system does, what data it uses and how its output influences decisions?

Could a customer, employee, regulator or executive reasonably understand the logic?


3. Consistency

  • Does the system apply understandable and predictable rules?

  • Are similar situations being handled in similar ways?


4. Control

  • Where does the data go?

  • Who can access it?

  • Which vendors, models or third parties process it?

  • Can the organization intervene when required?


5. Consequences

  • Who benefits?

  • Who could be harmed?

  • And what happens when the system is wrong?


Consider an AI system that recommends customer pricing based on demand, customer history, equipment availability, geography and payment behavior. The financial opportunity may be obvious, but leadership still needs to ask:


  • Can we explain why two customers received different prices?

  • Can we explain which data influenced the recommendation?

  • Does the AI provider retain our information?

  • Could the model unintentionally disadvantage a particular group of customers?

  • Who has the authority to override the recommendation?


If leadership cannot answer those questions clearly, the organization may be accepting risks it does not fully understand.


Finance Has a Bigger Role to Play


Finance should have a much larger role in AI than simply calculating return on investment.

Finance already understands controls. We understand assumptions. We understand accountability, materiality, risk, governance and decision rights. Those disciplines should be built into AI initiatives before the technology is deployed, not after something goes wrong.


One simple place to start is an AI register.


For every meaningful AI use case, document:

  • The business owner

  • The technology or vendor being used

  • The data being processed

  • The decision or process being supported

  • The employees, customers or stakeholders affected

  • The risks involved

  • The person with authority to intervene


This is not sophisticated and that is precisely why it works. Before organizations build elaborate AI governance structures, they first need visibility and ownership.


AI Will Reveal Your Leadership

AI can forecast. It can automate. It can analyze. It can communicate. It can help organizations make better decisions, serve customers more effectively and create substantial economic value.


But it cannot decide what your organization should stand for. It cannot decide what is fair.

It cannot take responsibility when a customer is misled, an employee is treated unfairly, confidential information is exposed or a poor decision is made. Those responsibilities remain with leadership.


The real question is no longer whether organizations will use AI -- Most already are.

The question is whether leadership, governance and culture are advancing as quickly as the technology.


AI will not save your organization


Leadership will determine whether AI strengthens it or undermines it.

So, take one AI initiative in your organization and put it through the Five Cs:


Consent. Clarity. Consistency. Control. Consequences.


If you cannot answer the five Cs clearly, the technology may not be ready. Your organization may not be ready to scale it.

------------------------------------

About the Author

Achille Ettorre is the author of The Digital Advantage and an executive, operator and advisor focused on AI, analytics, digital transformation and business performance. His work focuses on helping leaders turn technology into better decisions, stronger operating discipline and measurable business outcomes.

 
 
 

Comments


bottom of page