The conversation about AI and consulting has been framed incorrectly from the start.
The dominant narrative — that AI tools are making consultants redundant by doing their work faster and cheaper — is both partially true and almost entirely beside the point. The consultants who are being displaced by AI are being displaced because AI can now do something they were already doing poorly: synthesizing publicly available information, generating slide frameworks, and producing first-draft analyses that look rigorous but rarely are.
That work was always the weakest part of the consulting value proposition. AI has not destroyed it. It has made its weakness visible.
What AI actually does to consulting
Clients are acquiring AI tools. This is happening at every level of organizational sophistication, from large enterprises building internal AI capabilities to mid-size companies using off-the-shelf tools that would have required a consulting team to produce five years ago.
What these tools give clients is analytical capability — the ability to model more scenarios, synthesize more information, and generate more options in less time. That is genuinely valuable and genuinely reduces the need for certain kinds of consulting work.
What these tools do not give clients is a decision framework. They do not tell the organization which of the scenarios to choose, on what basis, with what trade-offs explicitly made. They do not produce a documented rationale that survives leadership change or regulatory audit. They do not provide a governed process for turning analytical output into accountable decisions.
The gap between analytical capability and decision quality — always present, rarely discussed — is widening as AI tools proliferate inside client organizations. More analysis. Same ad hoc decision processes. The same expensive mistakes, now better-informed.
What this means for the consulting industry
The consulting firms that are most exposed to AI disruption are those whose model depended on information asymmetry — on knowing things the client did not know, or being able to synthesize information the client could not access or process quickly enough.
That asymmetry is gone, or going. A client with good AI tools and a capable internal team can now access and synthesize information that previously required a consulting engagement to produce. The analysis is not always as good — but it is good enough for many purposes, at a fraction of the cost and time.
The consulting work that is not exposed — and may actually grow in demand as AI tools proliferate — is the work that was never really about information. It is the work of building decision governance into organizations. Of taking the analytical capability the client now has and creating a structured framework for turning it into decisions that are explicit, defensible, and documented.
An organization that has invested in AI analytical tools but has not invested in decision governance has better inputs and the same broken process. The gap between what the analysis says and what the decision actually reflects — between the model’s recommendation and what happened in the meeting room — is as wide as it ever was. AI has not closed it. If anything, it has made it more visible.
The value that remains
Good consulting has always been about two things that AI does not provide: intellectual architecture and implementation discipline.
Intellectual architecture is the methodology — the structured approach to a class of problems that is grounded in theory, validated by experience, and adaptable to context. It is not information. It is a framework for using information to make better decisions. That framework takes years to develop and cannot be replicated by a tool that has no stake in whether the decision it supports turns out to be right.
Implementation discipline is the work of taking that methodology and actually embedding it in an organization — configuring it to the specific competitive environment, building the habits and processes that make governance real rather than nominal, and calibrating the framework against actual outcomes over time. That work requires human judgment, organizational understanding, and the kind of ongoing relationship that a software tool does not sustain.
These are not small things. They are exactly what the organizations that navigate AI disruption successfully will invest in — and what the organizations that mistake better analysis for better decisions will lack.
What clients should actually be asking
The question most organizations are asking about AI and consulting is: what can we now do ourselves that we used to pay consultants to do?
That is a reasonable question. But it is the wrong question for the moment.
The right question is: now that we have better analytical capability, what decision framework are we using to turn that capability into better decisions? Who challenges the assumptions our analysis produces? How are the trade-offs explicitly made? What is documented when a decision is reached, and what happens when it is overridden?
If the answer to those questions is still “we discuss it in a meeting,” the organization has better analysis in service of the same broken decision process. AI has not helped. It has just made the dysfunction more sophisticated.
What this means for consultants
The consulting firms — and the independent practitioners — who will thrive in the AI era are those who can answer that question concretely. Not with a framework that was relevant five years ago and has been rebranded. With a genuine decision governance capability that turns the client’s AI-enhanced analytical output into something an organization can actually be accountable for.
That is not a technology problem. It is a methodology problem. And methodology, grounded in experience and validated by outcomes, is exactly what good consulting has always been for.