Interview: Bain partner on what AI changes in the job market

This piece first appeared in German. The English version is a rewrite rather than a line-by-line translation, and the answers below are translated: the substance is unchanged, the wording is not the speaker's own. The German original stays online in the archive: Interview mit Bain Partner: Wie verändert KI den Arbeitsmarkt?.

All answers are the personal assessments of the interviewee and are not to be read as an official statement by his employer.

A retail credit decision at a Spanish commercial bank used to take weeks or months, with manual checks and hierarchical sign-off at every stage. An AI-supported system brought the average down to 17 minutes. Axel Erhard, a senior partner at Bain, offers that case as his example of what the current wave of adoption actually does inside a company. The conversation below covers what it means for the people working there.

Where the change lands first

Which areas of a company do you see changing most in the short term, and where will the human role resist automation longest?

AI brings more than efficiency gains in the sense of faster and with fewer resources. It brings clear gains in effectiveness: processes get qualitatively better.

Take lending in banking. A classical process at a Spanish commercial bank took weeks or months for retail customers and required numerous manual checks along with hierarchical approval steps. An AI-supported system reduced processing time to an average of 17 minutes.

The gains come from the analytical capabilities: more consistent and more objective risk assessments, fact-based optimization, and better pricing, because borrowers carry different risk profiles. One price fits all no longer belongs to this era. For banks that means headcount falls in the branch network while head offices need more specialists to monitor the AI-supported systems and keep improving them.

And the roles that stay human?

Some areas will not automate quickly. Technological innovation still has to be managed and developed by people. Occupations with a strong human element, in care, education and health, remain indispensable. Wherever physical goods are produced outside highly automated processes, people are still needed.

From enthusiasm to durable value

Many companies are investing in AI at the moment, often out of enthusiasm for the subject itself. How can a firm make sure AI produces long-term value and not only short-term efficiency?

The excitement often leads to experiments with no clear objective. AI must not be introduced for its own sake; it has to follow a clear business logic.

Because the environment changes far faster than it used to, a static ten-year strategy is no longer enough. Companies need agile and adaptive strategies. The successful ones start small cloud-based pilots, integrate them, and build a coherent AI ecosystem step by step.

What leaders do with their time

How does the role of a manager change when decisions are increasingly prepared or proposed by data?

Leaders should be investing around 80 percent of their time in vision and strategy, and only 20 percent in operational management. AI now delivers nearly all the relevant information on demand, so the value of a good leader lies more in setting direction and developing a picture of the future.

Lost jobs or new roles

Many industries worry about job losses. Do you think AI destroys jobs, or creates new roles?

The central question is which capabilities will be needed over the long run. A few years ago the software engineer was the future-proof profession; today many people talk about the technology engineer. What matters is openness to technology, willingness to learn, and the ability to integrate new tools into processes quickly.

AI will play a part in almost every occupation, and anyone who does not take it up risks being left behind. Even a highly specialized neurosurgeon will achieve worse outcomes in the medium term by ignoring AI-supported diagnostics that detect tumors more reliably.

The consulting industry itself

How is AI changing consulting, particularly at the Big Three?

Consultancies will play a key role in guiding clients through the use of AI and helping them build a competitive advantage from it. The classical pyramid with its broad base of business analysts will erode, because entry-level tasks such as data analysis or due diligence can be done faster, more precisely and more objectively by machine.

New roles appear at the same time: technically minded consultants who are not pure coders but understand what AI makes possible and how to translate that into a specific client solution. Much like the industrial engineers of an earlier generation, who connected engineering with commercial reality, the field now needs tech-savvy consultants who can bridge technology and business.

A note on McKinsey

The shift is visible at McKinsey. Business Insider reported in May 2025 that the firm's headcount had fallen by more than ten percent in eighteen months, from about 45,000 to about 40,000. McKinsey attributed the decline to normal attrition and performance reviews, not to AI.

What AI is changing is the work inside the pyramid. McKinsey's internal platform Lilli now drafts proposals and builds slides from simple prompts, and more than 75 percent of the firm's employees use it every month. "Do we need armies of business analysts creating PowerPoints?" asked Kate Smaje, its global leader of technology and AI. "No, the technology could do that."

Even the largest consultancies are exposed to the disruptive force of this technology. For the industry it means a rebuild of structures, away from the traditional pyramid with its broad analyst base, toward leaner models where technical competence and the ability to translate it strategically are what count.

That has to start in education, so that enough qualified people exist on the other side of the change. Degrees and training in industrial engineering, business informatics or data science with a management component are worth considering, as are specialized programs in digital business, AI and technology management. Alongside them, further training in agile project leadership, cloud technologies and applied AI will do much of the work of producing the next generation of tech-savvy consultants.

Sources

Grouped by the section they support.

Opening

• No external source. The lending case at a Spanish commercial bank is the interviewee's own example; the bank is not named.

Where the change lands first

• No external source. The assessments are the interviewee's own.

From enthusiasm to durable value

• No external source. The assessments are the interviewee's own.

What leaders do with their time

• No external source. The 80/20 split of a leader's time is the interviewee's own view, not a measured figure.

Lost jobs or new roles

• No external source. The assessments are the interviewee's own.

The consulting industry itself

• No external source. The assessments are the interviewee's own.

A note on McKinsey

• Merritt (2025), McKinsey's headcount drops more than 10% in 18 months, Fortune with Morning Brew, 29 May 2025, fortune.com, following a report by Business Insider. Supports the fall from about 45,000 to about 40,000 employees and McKinsey's explanation of attrition and performance reviews.
• El Chmouri (2025), McKinsey leans on AI to make PowerPoints faster, draft proposals, Fortune with Bloomberg, 2 June 2025, fortune.com. Supports what Lilli does, the more than 75 percent of employees using it monthly, and Kate Smaje's remark on business analysts and PowerPoints.
• The conclusions for education and training are the interviewee's own.