The firm that teaches CEOs how to manage AI is now learning what happens when AI manages the firm.

The Summary

The Signal

McKinsey is rebuilding itself with AI while watching its clients fumble the same transformation. Bob Sternfels sat down with Bloomberg as Nvidia faced a leadership crisis, which makes his comments on corporate resilience hit different. When the company powering the AI revolution stumbles on succession planning, the consultants taking notes have permission to say the quiet part loud.

Sternfels discussed how AI is reshaping McKinsey's workforce, which is consultant-speak for "we're automating the junior analyst work that built careers for decades." The firm that wrote the playbook on organizational efficiency is now stress-testing that playbook on itself. This matters because McKinsey doesn't experiment. When they move, it's because the math works.

"What CEOs are getting wrong about adopting the technology" isn't just a soundbite—it's McKinsey's admission that the C-suite is lost.

The pattern emerging: companies know AI matters, they're throwing money at it, but they're missing the fundamentals. Strong governance. Clear succession plans. The boring infrastructure that makes crisis-proofing possible. Nvidia's leadership drama is Exhibit A. You can have the most valuable AI chips on the planet and still trip over organizational basics.

McKinsey's actual transformation tells a different story than their client engagements. Internally, they're testing:

  • Which knowledge work tasks agents can fully own
  • How to restructure teams when AI handles the grunt work
  • What their workforce looks like when junior roles evaporate

Sternfels also flagged vulnerabilities in global trade, which connects to AI in ways most coverage misses. Trade fragmentation makes AI supply chains fragile. Semiconductor geopolitics isn't abstract policy when your entire AI strategy depends on Taiwan staying stable and ASML export licenses getting approved.

The Implication

Watch what McKinsey does, not what they say in client presentations. If they're restructuring their workforce around AI, your company is next. The consultants optimizing themselves out of entry-level jobs today are previewing what every white-collar profession faces tomorrow.

For CEOs: if you're thinking about AI as a technology problem, you've already lost. It's a governance problem, a succession problem, a "what does this company look like in three years" problem. The technology is the easy part. Nvidia proved that. Building organizations that can absorb this level of change without imploding is the actual work.

Sources

Bloomberg Tech