They should not be treated as interchangeable. In the avatar pilot, AI plays a simulated client that you converse with. Lilli is McKinsey’s internal AI platform, and separate reporting has described candidate exercises involving that platform.

A July 2026 description of an AI-assisted case also discussed use of an AI tool during part of case work. Public sources have not completely clarified how those case-interview pilots relate to reports of Lilli-related exercises, so it would be too strong to claim they are either identical or wholly separate standardized assessments.

A practical distinction is the role AI plays. Talking to a simulated client tests communication in a conversation, while using an AI tool during analysis calls for checking AI output and exercising independent judgment. These are descriptions of the reported formats, not official grading categories.

Check which exercise appears in your recruiting instructions before you prepare for a particular workflow. Our guide to McKinsey Lilli explains the platform and the reported interview context.

For a deeper explanation, see our AI avatar interview guide for a clearer comparison of the formats.

The discussion of what happens during the AI avatar conversation helps illustrate why it differs from working with an AI analysis tool.

That explains why some summaries seem to mix the tools together.

I had assumed anything described as an AI case was automatically Lilli, which clearly is not established.