Consulting has a productivity problem hiding in plain sight. Highly trained people can spend hours finding sources, cleaning data, turning analysis into charts and adjusting slides. Those tasks matter, but they are not usually why a client pays for advice.
AI is changing the cost and speed of that preparation. It is also making it harder for consultants to justify their value through the volume of material they produce.
The consulting work AI is already affecting
Desk research, interview synthesis, drafting market overviews, structuring issue trees and creating first-pass slide narratives can all be assisted by AI.
That does not mean a generated market assessment is ready for a chief executive. The sources may be weak, the framing may miss the decision and the recommendation may not survive a sensible challenge.
A report can look like a consulting deliverable while containing no original insight at all.
The client is paying for a decision, not a stack of slides
Good consulting starts by defining the problem. Why is the client asking this question? Which decision must be made? What evidence would change the answer? Who needs to agree?
AI can help explore options, but it has no direct stake in the client's outcome. It cannot substitute for understanding the personalities in the room, obtaining access to the right data or owning the implications of advice.
The strongest consultant is not necessarily the person who generates the most content. It is the person who turns messy information into a defensible choice and gets the organisation to act.
How junior consulting roles could change
The analyst traditionally learns by doing research, structuring spreadsheets and preparing slides. As the easier parts become quicker, managers may expect junior consultants to reach judgement earlier.
That is an opportunity if teams teach the fundamentals. It is a risk if juniors become editors of AI-generated outputs without understanding the underlying method.
The future consultant needs both: the ability to work with AI and the ability to explain, from first principles, why the conclusion is right.
An exercise you can try on a live work pattern
Use only approved, non-sensitive material. Take a narrow public question, such as which of three competitors is best positioned for a particular customer group.
- Define the decision and the criteria before opening an AI tool.
- Ask for an evidence map, with primary sources and unresolved questions.
- Independently inspect the most important sources and reject unsupported claims.
- Build your own comparison of the alternatives.
- Draft a two-slide recommendation: the answer and the reasons.
- Ask what evidence would make you change your mind.
The purpose is to use AI to shorten preparation, not to outsource critical thinking.
Which capabilities matter more now?
Problem framing, quantitative rigour, commercial judgement, facilitation, leadership and client trust. These are not glamorous additions to a prompt-writing course. They are the work that turns research into an outcome.
Microsoft's 2026 Work Trend Index highlights how AI users increasingly see quality control and critical thinking as essential. Consulting should take that seriously. An elegant chart is not a substitute for a correct answer.
A useful companion is our guide to AI's effect on banking roles, followed by the practical 30-day plan.
