Last Updated on February 14, 2026
Between efficiency promises and reality
Artificial intelligence is widely regarded as the most significant transformation of consulting work in recent decades. Within just a few years, large language models have entered everyday project work, from market analyses and presentation drafts to internal knowledge databases. Hardly any other technology has been rolled out so quickly and so broadly within consulting firms. Expectations are correspondingly high: greater speed, lower costs, and smaller teams.
At the same time, a provocative claim persists: “AI will replace junior consultants.”
After all, new tools now perform exactly the kinds of tasks that traditionally defined the entry level into consulting: research, data collection, structuring information, initial analyses, and slide drafting. On paper, this suggests a structural reduction of junior roles, which in the medium term is difficult to dismiss entirely.
Yet today’s reality is far more complex.
Early internal experiences show that the division of labor, learning curves, and quality control are indeed changing, but not according to the simple logic of “humans out, machines in.” Instead, new dependencies, new risks, and new requirements for young consultants are emerging.
Against this backdrop, a central question arises: How is artificial intelligence actually changing recruiting, training, and the daily work of junior consultants, beyond hype and headlines?
AI in everyday consulting: new tools, new behaviors
Large language models are now a fixed component of project work in almost all major consulting firms. Access to powerful AI tools has been rolled out broadly, often through internal platforms connected to corporate databases, knowledge systems, and security layers. For project teams, this means that information, analyses, and text or slide drafts are available in seconds at a speed that would have been unthinkable just a few years ago.
Increasingly, it is not just individual AI tools that are used, but entire AI agents working alongside human consultants. The idea that future consulting work will be organized as collaboration between humans and AI agents is now shared at the highest leadership levels. McKinsey’s Global Managing Partner Bob Sternfels already describes the firm as a workforce of “human and agentic team members” and expects that in the near future every consultant will be supported by multiple AI agents.
Working with AI agents is becoming as natural as managing analysts or research teams today. These agents take on clearly defined tasks, deliver iterative results, and are steered much like human team members. For junior consultants, this creates a new reality: they use AI agents in projects in much the same way they themselves are guided by experienced project leaders, as executing, accelerating units within the team.
However, internal reports from several top consulting firms reveal new behavioral patterns, particularly at the junior level. AI is often used as the primary research and production tool, with a strong reliance on copy-and-paste workflows. The result is faster output, but frequently with less deep understanding. Where previously independent structuring, execution, and multiple iterations were required, the machine now takes over. As a result, deeper intellectual engagement is partly shifted or even diminished.
In parallel, many firms are seeing a decline in traditional internal research teams. Tasks that were once handled by specialized researchers can now often be covered directly by AI systems.
This new workflow introduces a central risk: fundamental consulting skills are being developed more slowly or less robustly. These include the ability to structure problems clearly, to manage different stakeholders such as experts, analysts, or researchers, to critically assess the quality and origin of sources, and to develop business judgment, meaning the capacity to distinguish plausible from implausible results under uncertainty. I also observe this trend when coaching candidates for case interviews: many have become accustomed to outsourcing core structuring skills to tools like ChatGPT. If you want to build your case interview skills from the ground up and stand out in this recruiting market, you need to develop problem-solving skills, analytical thinking, and top-down communication – all topics we teach in our Case Interview Academy.
The tools have become faster and more powerful. The question is whether thinking and awareness have evolved at the same pace.
Conversations with active consultants at leading firms suggest that a growing gap is emerging here, and that firms must actively counteract it.
New approaches in recruiting: the AI interview
One visible sign of this shift is already appearing in recruiting. Consulting firms are beginning to test new capabilities that are increasingly relevant in AI-supported work environments. A concrete example of this change is McKinsey’s currently piloted AI interview format.
Unlike classic digital assessments or asynchronous video interviews, this is not an automated interview. Instead, candidates engage in an interactive conversation with McKinsey’s AI system “Lilli,” accompanied by a human interviewer acting as a facilitator. The goal of the pilot is to understand how candidates interact with AI: how they prompt, ask follow-up questions, handle uncertainty, and critically challenge AI-generated content.
The format is currently being tested in selected North American and European offices, complementing established case and fit interviews. The focus is less on assessing traditional interview skills and more on observing new competencies that are becoming increasingly relevant in consulting: structured communication with AI systems, precise task formulation, iterative steering of outputs, and the ability to contextualize and validate results.
For candidates, this means that alongside problem-solving, analytical, and communication skills, confident handling of AI as a working tool is becoming more important.
McKinsey is a pioneer here, but experience suggests it will not take long for other firms to follow suit, both to remain competitive in the talent market and to signal externally that they are actively shaping the transformation of consulting work rather than merely observing it.
Productivity gains: why they are smaller than expected
With the arrival of powerful AI systems came a clear expectation: project teams could become smaller while delivering the same quality and speed. Especially in consulting, where a significant portion of work involves research, analysis, and presentation development, the potential for drastic efficiency gains seemed obvious.
However, early experience paints a more nuanced picture.
The real challenge rarely lies in the technology itself, but in redesigning the working models it enables and requires. Organizational structures, decision pathways, and responsibilities must be rethought. This transformation is a prerequisite for sustainable productivity gains and takes time, which explains why progress has been slower than many initially predicted.
As a result, team sizes on many projects have remained largely stable. Instead of using fewer people, the objective has shifted toward producing more output in the same time. More analyses, deeper scenarios, more client iterations, and more refined recommendations have become possible. Expectations have risen, but long working hours remain unchanged.
At the same time, new time sinks are emerging. AI-generated outputs must be validated, sources checked, and numbers recalculated. Because large language models can produce plausible-sounding but incorrect content, the workload for fact-checking has increased significantly. Moreover, project teams must increasingly judge whether an AI result is truly correct or merely convincingly formulated.
The net effect is therefore clear: efficiency gains are real and noticeable, but so far they fall well short of the hype predictions that promised massive short-term productivity and cost reductions. AI accelerates work, but it does not eliminate the need to understand, contextualize, and take responsibility for results.
A changing learning model for new hires
For a long time, entry into consulting followed a proven learning model.
Junior consultants developed their skills primarily through hands-on work: independently defining and structuring problems, analyzing data, and iteratively preparing content. Gathering, organizing, synthesizing, and analyzing information was not just a means to an end, but a central part of professional development. It was during these early project phases that the foundation for later problem-solving ability and business judgment was built.
With the widespread use of AI, this model is shifting fundamentally. Portions of execution — from information search to initial analyses and text or slide drafting — are increasingly handled by AI systems. This increases speed but reduces the “friction” through which learning traditionally occurred.
This creates a clear risk: when AI replaces steps that previously served as training ground, learning is partially skipped. Skills such as clean structuring, critical source evaluation, or recognizing inconsistencies develop more slowly when consultants are less exposed to raw material. At the same time, senior leaders emphasize that core human capabilities are becoming more important, not less. Internal discussions often highlight judgment, ambition setting, and creative, non-linear problem-solving as capabilities that AI systems cannot structurally replicate.
For consulting firms, this creates a new challenge. Traditional “learning by doing” models are no longer sufficient. Instead, deliberately designed training formats are needed to build foundational skills in a targeted way — alongside, not after, the use of AI. The question is not whether AI should be used, but how to ensure that junior consultants continue to learn how to think independently.
What does this mean for young consultants?
For aspiring consultants, expectations in the recruitment process are not changing through a complete criteria overhaul, but through a shift in the mix of skills. Classic core competencies such as structured problem-solving, rigorous analysis, clear communication, and professional presence remain central selection criteria. At the same time, it is becoming more visible who can effectively steer AI as a tool, critically validate results, and transparently communicate uncertainty.
Critical thinking has always been essential in consulting. What is new is that AI can generate convincing but flawed or incomplete results at high speed. At the same time, AI drastically lowers the cost of producing output, increasing the risk of unverified, redundant, or incorrect work. Consequently, the ability to test assumptions, identify contradictions, and methodically assess AI outputs is becoming more important. It is not the speed of prompting that matters, but the quality of steering and subsequent evaluation. Candidates who use AI selectively, consistently verify results, and openly acknowledge limitations send a strong signal of quality.
In line with these changes, junior consultants must manage growing complexity, coordinate multiple workstreams in parallel, and steer AI agents that support them in project work.
Consulting remains a people business. Building trust, translating complex content into understandable language, and mediating between human judgment, machine outputs, and client decisions will continue to gain importance. The interface between human, machine, and client is becoming a critical success zone.
In recruiting, this therefore represents less a replacement of existing interview formats than an expansion to include new signals. Case and fit interviews remain central components. In addition, firms are increasingly assessing whether candidates understand when AI is useful, when it should deliberately not be used, and how they handle uncertainty and responsibility.
Learning ability, resilience, and collaboration are coming more into focus. Leading firms are systematically analyzing which traits define long-term successful consultants and adjusting their assessments accordingly. The common thread remains: what matters is not finished knowledge, but the ability to continuously adapt in a rapidly changing work environment.
Outlook: the role of junior consultants remains but changes
Artificial intelligence will not replace junior consultants in the short term. In the medium term, however, profound changes are likely here as well. This applies to consulting just as it does to many other knowledge-intensive industries and professional services.
Currently, even at top management level, there is no talk of eliminating entry-level roles, but rather of shifting their profile. Consulting work is moving away from standardized analysis toward more complex, interconnected questions where human judgment, creativity, and responsibility remain indispensable.
Consulting remains a business built on human judgment, accountability, and relationship-building. High-stakes decisions cannot be delegated to machines alone, and clients continue to expect people who can contextualize results, take responsibility, and build trust.
At the same time, the classic learning curve in consulting is being redesigned. Many tasks that previously served as natural training stages for junior consultants are now accelerated or partially automated by AI. As a result, the path from first assignment to independent problem-solving becomes shorter but also more demanding: less repetition, more responsibility, and the need to develop complementary skills earlier in one’s career.
