McKinsey & Company, the esteemed global consulting firm, has recently launched their generative AI tool known as "Lilli"^[1][2].

WHY LILLI?

Named in honor of Lillian Dombrowski, the first professional woman hired by McKinsey in 1945, Lilli carries the torch of her legacy by streamlining the process of accessing the firm's vast knowledge repository. This transformative tool is designed to empower individuals within McKinsey by unifying their resources and capabilities. Lilli aims to liberate more time for teams to focus on problem-solving, coaching, and assisting clients in realizing their performance aspirations^[1].

LILLI'S FUNCTIONALITY

Lilli's operation relies on more than 40 McKinsey-curated information sources, and over 100,000 documents and interview transcripts, drawing on the expertise of a vast network that spans 70 countries. Having this extensive range of knowledge at its disposal, Lilli can create initial project plans which used to take senior colleagues weeks of researching and networking, in a much more efficient manner^[1].

What sets Lilli apart is its ability to identify and make connections between different areas in projects to identify argument weaknesses and anticipate questions that may arise. This process optimization could save up to 20% of time that would typically be spent researching, planning and preparing for meetings^[1][2].

USER INTERFACE AND FEATURES

Lilli's user interface includes a text-entry based chat where responses are generated in a chronological order, allowing employees for a seamless interaction with the platform. It has additional features like a sidebar with saved prompts for enhanced utility. McKinsey's ClienTech team, led by CTO Jacky Wright, ensured that the platform was categorized into two tabs: The “GenAI Chat” extracts information from a generalized language model (LLM) backend, and “Client Capabilities” sourcing responses from McKinsey's vast internal corpus^[2].

THE FUTURE OF LILLI

Since its beta release in June 2023, Lilli had been used by approximately 7,000 employees, who reported a drastic reduction in research and planning work, from weeks to hours and in some instances, from hours to minutes. By the end of 2023, McKinsey hopes to increase Lilli’s user base to thousands, following additional prompt engineering, content validation, and more upskilling measures^[1][2].

Considering its success within the firm, McKinsey envisions Lilli finding applications in every consultants' work, from initial research to project implementation. Moreover, the firm is also exploring the possibility of extending Lilli as an external-facing product for clients or other firms^[2].

CONCLUSION AND KEY TAKEAWAYS

McKinsey’s innovative endeavor to harness the power of AI shines a new light on the potential benefits of technology in corporate settings. The company’s generative AI tool, Lilli, has demonstrated how AI can optimize internal workflow, drive efficiencies, and save significant research and planning time. The tool is likely to redefine the way firms explore their vast knowledge landscapes in the future.

Moreover, the contemplation of making Lilli an external-facing tool underscores its potential uses beyond McKinsey’s immediate organizational needs. This example sets an interesting precedent for other businesses to follow, signifying the growing momentum of AI's application within the corporate arena.

Sources

[1] "Meet Lilli, our generative AI tool that’s a researcher, a time saver, and an inspiration". McKinsey, 2023. Link
[2] "Consulting giant McKinsey unveils its own generative AI tool for employees: Lilli", Carl Franzen. VentureBeat, 2023. Link