In recent months, uncomfortable stories have emerged across the consulting industry. Companies have charged hundreds of thousands of dollars for strategic reports, some of which were produced with extensive use of Artificial Intelligence tools. In some cases, these reports contained significant errors, weak recommendations, and financial and reputational consequences for the consultants who produced them.
The reaction was immediate and one of outrage. But perhaps the more honest question is a different one. Is this really so different from what has always happened?
Consulting has always been a school, one of the best
For decades, the consulting model has followed a relatively simple logic: teams with a strong contingent of junior consultants, recent graduates with little or no experience, carry out the analytical work, which is then reviewed by senior consultants. These less experienced professionals build models, analyze data, research benchmarks, structure presentations, and produce large portions of consulting reports.
Inevitably, they make mistakes. Not because they lack competence, but because they lack experience, which is only natural at the beginning of a career. The system has always worked this way because it has an important strength: this is how consultants learn. It is through this process that they develop critical thinking, analytical skills, structured reasoning, and, above all, the ability to ask the right questions. These are precisely the qualities that later distinguish an excellent senior consultant. If this has always been the model, then the question becomes: what really changes with the arrival of AI?
AI does not replace consultants. It replaces tasks.
Artificial Intelligence has accelerated work that already existed. Today, AI tools can accomplish in minutes what once took hours or even days: synthesizing information, structuring reports, suggesting frameworks, analyzing data, and producing preliminary recommendations. This raises an obvious question. If a significant portion of a junior consultant’s work can be automated, what role remains for junior consultants? Perhaps the more important question is this: what is the real value of consulting? The value has never been the report itself. A report, by itself, has never had value. The value lies in the confidence, applicability, and impact that come from review by someone with greater experience. Clients do not pay for slide decks. They pay for better-informed decisions. They pay for practical solutions. They pay for the ability to reduce risk and increase the likelihood of success. And that has never depended solely on the person who produced the document. It has always depended on the person who thought it through.
The truth is simple: a poor report, whether produced by an AI model or by an inexperienced junior consultant, suffers from the same problem. It adds no value. The difference is that AI makes the problem more visible. Responsibility remains human.
The recent cases of reports containing serious errors expose more than technological shortcomings. Above all, they expose failures in governance. It has never been, nor should it ever be, the responsibility of a junior consultant to guarantee the overall robustness of a project. That responsibility has always rested with senior teams. AI does not change that logic, but it does make it more demanding.
What about the future?
The invisible risk that I do not see anyone discussing is this: how do we train the consultants of the future?
If AI replaces a significant portion of the work performed by junior consultants, how do we develop tomorrow’s senior consultants? The traditional career path was built on practical learning. It was through this process that the profession’s greatest asset was developed: critical thinking.
If we eliminate or drastically reduce this stage, we run a real risk of creating a generation that knows how to use tools but does not know how to interpret results. That distinction is critical. Using AI is not about knowing how to ask for answers. It is about knowing how to ask the right questions.
Learning cannot be automated
A dangerous illusion is beginning to take hold: the belief that efficiency can replace learning.
AI can accelerate processes, but it cannot replace the development of reasoning. It cannot replace experience. It cannot replace exposure to real-world problems. A consultant who has never analyzed data in depth will struggle to question an AI-generated analysis. A professional who has never made mistakes will struggle to recognize them. And this applies not only to consulting, but to every knowledge-based profession.
The real risk is not that AI does too much. The real risk is that people stop learning enough.
The model that will endure
Consulting is not going away. It will adapt.
The most successful models will be those that strike the right balance between four factors: technological efficiency, human development, experienced supervision, and robust outcomes.
The firms that will lead this transformation are not necessarily those that replace people with AI. They are the ones that know how to integrate AI without sacrificing critical thinking and continuous learning.
This means continuing to invest in junior consultants, but with different responsibilities: less mechanical production, more interpretation, greater involvement in decision-making, and more interaction with clients.
Above all, it means reinforcing the role of senior consultants as the ultimate owners of every recommendation, just as they have always been.
The question is not whether AI is replacing consultants. The question is whether we are using AI to improve consulting or to conceal weaknesses that have always existed. Because the qualities that truly create value have never changed: analytical ability, critical thinking, experience, and accountability.
AI may produce reports more quickly. But it cannot, and will not, replace expertise or genuine added value. And that is precisely why the future of consulting will not depend on who uses technology most effectively. It will depend on who continues to think more effectively.
Rute Xavier, Professor at Católica-Lisbon SBE