Why The “AI Is Replacing Consultants” Narrative Gets It Wrong, And What Actually Needs To Happen Next

On March 1, 2026, Entrepreneur published an article titled “Consultancy Is Dying Thanks to AI,” arguing that Gen Z should skip consulting entirely because AI can do the work cheaper and faster. The author went so far as to compare consultants to floppy disks.

The argument sounds compelling on the surface, but it fundamentally misunderstands what consulting actually is, what clients actually pay for, and where the real risk in AI adoption lives.

The real problem is not that AI is replacing consultants. The real problem is that organizations are deploying AI without the literacy, governance, or human oversight to use it well. And we have credible data from McKinsey, Gartner, and Stanford that proves it.

When AI Gets It Wrong, The Damage Is Already Done

In 2024, Air Canada’s AI chatbot told a grieving customer he could apply for a bereavement fare In 2024, Air Canada’s AI chatbot told a grieving customer he could apply for a bereavement fare discount within 90 days of travel. He relied on that information, booked full-price flights to attend his grandmother’s funeral, and later submitted a refund request. Air Canada denied it. Their actual policy did not allow retroactive bereavement claims.

When the customer sued, Air Canada argued the chatbot was a “separate legal entity” that could not bind the company. The British Columbia Civil Resolution Tribunal rejected that argument entirely and held that a company is responsible for all information on its website, whether from a static page or a chatbot. Air Canada was found liable for negligent misrepresentation.

That same year, in a New York federal court, attorneys representing a plaintiff in Mata v. Avianca submitted a legal brief citing six judicial opinions. Every one of them was fabricated by ChatGPT. The AI invented case names, generated fake quotes, and created fictional citations. The judge sanctioned the lawyers $5,000. The case is now the leading example of AI hallucination consequences in the legal profession.

These are not outliers. McKinsey’s 2025 State of AI report, surveying nearly 2,000 executives across 105 countries, found that 51% of organizations using AI have experienced at least one negative consequence, with inaccuracy as the most commonly reported issue. Stanford’s 2025 AI Index recorded 233 AI-related incidents in 2024 alone, a 56.4% increase over 2023. AI does not know it is wrong. It delivers fabricated information with the same confidence as verified facts. And when nobody in the room has the judgment to question the output, companies act on fiction.

What AI Cannot Do, And Why It Matters More Than Ever

Here is what AI cannot do. It cannot feel the tension in a room when a CEO is about to make a decision they have not fully thought through. It cannot call a client at 10pm on a Sunday because it sensed something was off in the last board meeting. It cannot read the body language of a founder who says “we’re fine” but means “we’re three months from running out of cash.” It cannot build the kind of trust that takes years to earn and seconds to lose.

That is not a small limitation. That is the entire business.

The Entrepreneur article defines consulting as frameworks, slide decks, and process optimization, then points out that AI handles those tasks well, and concludes the profession is obsolete. That is like saying restaurants are dead because vending machines can dispense food. It confuses the delivery mechanism with the value.

Yes, the commodity layer of consulting is being automated. Benchmarking, data synthesis, market sizing, template deliverables. That work is being absorbed by AI and it should be. It was never the valuable part. The value lives upstream: in judgment, in relationships, in the ability to sit across from a leadership team in crisis and help them see what they cannot see themselves. In asking the question nobody in the room realized needed to be asked.

AI does not ask uncomfortable questions. Consultants do.

The Literacy Gap Is Where The Real Risk Lives

Gartner has been direct about this: unlocking AI’s business potential requires building enterprise AI literacy, and most organizations are nowhere close. Their research found that even in the best-case scenarios, with proper resourcing and enthusiastic stakeholders, there are gaps in the skills and knowledge needed to effectively and responsibly use AI. Organizations that prioritize AI literacy for executives will achieve 20% higher financial performance by 2027 compared with those that do not.

Gartner’s chief of research, Daryl Plummer, put it bluntly at the 2025 IT Symposium: “AI is stealing your skills.” Their strategic predictions warn that by 2026, critical-thinking skills atrophy from GenAI use will push 50% of global organizations to require AI-free skills assessments. Think about that. Half the world’s organizations will need to test whether their people can still think without a machine.

A Gartner survey of over 700 CIOs found that by 2030, zero percent of IT work will be done by humans without AI. Every single role will involve AI in some capacity. If that is the trajectory, then literacy is not optional. It is the foundation everything else sits on.

The Skeptics Belong At The Table, Not On The Sidelines

Every strong governance framework I have seen built started with someone willing to slow the room down and say, “Wait. What happens when this goes wrong?” The early adopters are great at speed. The skeptics are great at durability. You need both at the table.

McKinsey’s own data supports this. Their research found that high-performing AI organizations are more likely to have defined processes for determining how and when model outputs need human validation. The practice with the strongest correlation to AI value creation? Human-in-the-loop oversight. The companies getting the most value from AI are not the ones moving fastest. They are the ones that built the human oversight mechanisms before they scaled.

Companies that sideline the cautious voices in the name of speed are going to spend double their original budget cleaning up the mess later. I have watched it happen with digital transformation. I have watched it happen with cloud migration. And I am watching it happen right now with AI.

Efficiency Without Oversight Is Not A Strategy. It Is A Countdown.

Here is where the numbers get uncomfortable. A May 2025 Gartner survey of 506 CIOs found that 72% of organizations are breaking even or losing money on their AI investments. Not gaining. Not transforming. Breaking even or losing. That is not a technology failure. That is an implementation failure. It is what happens when companies buy the tool without building the capability to use it.

Meanwhile, only 6% of organizations in McKinsey’s survey qualified as high performers, meaning they captured meaningful enterprise value from AI. Six percent. Out of 88% adoption. That gap is not about the technology being immature. It is about companies skipping the human infrastructure that makes the technology work.

Gartner also predicts that by the end of 2026, “death by AI” legal claims will exceed 1,000 globally due to insufficient AI risk guardrails. The short-term savings look great in a board presentation. But the cost of cleaning up AI-driven decisions that nobody understood or verified? That is where the real number lives.

.

The Consultants Who Evolve Will Be The Most Valuable People In The Room

Consulting is not dying. But the consultants who refuse to learn AI, who pretend it does not exist, or who cannot adapt their value proposition? They are done. And the article that calls the entire profession obsolete is giving companies permission to make a dangerous mistake: replacing human judgment with machine output and calling it progress.

The consultants who learn to pair human judgment with AI capability, who can sit with a client, verify what the machine produced, catch the hallucination, read the room, and make the call that no algorithm would ever make? Those people are about to become indispensable.

The future does not belong to AI alone. It does not belong to consultants alone. It belongs to the people who understand both well enough to know when to trust the machine and when to trust themselves.

That is not a dying profession. That is the most important one in the room.