One thing McKinsey is making very clear to job applicants is that being comfortable with artificial intelligence is now a basic requirement, not an added benefit. For some candidates, using AI during interviews has become mandatory, and those who struggle to work with it may find themselves out of the running. The global consulting firm has started incorporating AI-based assessments in select final round interviews for graduate roles, especially in the US. As part of this process, candidates are asked to work directly with Lily, McKinsey’s internal AI tool. If an applicant cannot effectively use the tools to solve business problems, it can directly affect their chances of getting the job. Here’s everything we know about it.

Use of AI tools is mandatory in McKinsey job interviews: The whole story explained

According to CaseBasics, a US-based company that helps candidates prepare for consulting interviews, McKinsey has introduced “AI interviews” for some final-stage applicants. In this round, candidates are given practical consulting tasks and are expected to complete them with the help of Lily.

CaseBasics explained that the interview is designed to test how well candidates can work with AI rather than test technical expertise. The company said, “In a McKinsey AI interview, you are expected to prompt the AI, review its outputs, and implement decisions to provide clear and structured feedback. The focus is on collaboration and reasoning rather than technical AI expertise.”

During the interview, candidates usually receive a business question similar to actual client work. Instead of relying solely on their analysis, they are encouraged to use AI to gather information, organize their thinking, and improve the end result. The emphasis is on how candidates use the tool, not on whether the AI ​​gives the correct answer.

Importantly, applicants are not expected to be experts in AI prompting or to understand how such systems are built. CaseBasics notes that candidates simply need to demonstrate that they can use AI as a “productive thinking partner” and clearly explain how they reached their conclusions, just as consultants do when working with junior colleagues.

Initial feedback suggests that AI interviews focus on decision making, decision making and clear communication. As CaseBasics said, “McKinsey AI interviews appear to assess how candidates think, make decisions, and collaborate with AI tools rather than their technical AI knowledge.”

AI Interview does not replace the existing assessment round. It is conducted along with two other assessments. One tests problem-solving and structured thinking, while the other focuses on personal influence, leadership skills, and values.

Lilly’s use in McKinsey’s hiring process was first reported by the Financial Times. McKinsey has not commented publicly on the new interview format. However, company leadership has spoken openly about how deeply embedded AI is now in day-to-day work.

McKinsey now has 25,000 AI agents alongside human employees

McKinsey CEO Bob Sternfels has said that the company works with a large number of AI agents alongside human employees. In a recent episode of the Harvard Business Review IdeaCast, Sternfels said that McKinsey’s total workforce is about 60,000, of which about 20,000 are AI agents. A few days later, speaking at the Consumer Electronics Show in Las Vegas, he clarified that the real number was closer to 25,000. McKinsey later confirmed this figure to Business Insider.

This means that more than a third of the company’s workforce now consists of AI agents. Sternfels also noted how quickly it happened. Just a year and a half ago, McKinsey was using only a few thousand AI agents. The company now expects that within the next year and a half, every employee will be working with at least one AI agent as part of their daily tasks.