What the Companies Blaming AI for Layoffs Are Actually Telling You

The bottleneck on my current team is not writing code.

My engineers use AI to hunt down the source of bugs: pulling logs, tracing call stacks, surfacing the likely origin of a failure faster than any manual search would. What AI does not do is decide whether the proposed fix is the right one. Whether the solution is clean. Whether it solves the actual problem or patches the visible symptom while leaving the underlying issue intact. That judgment is what the engineers produce. AI accelerates the search. The engineers apply the decision.

That is what augmentation looks like when the operating system is already there.

On May 19, 2026, KPMG and Anthropic announced a global strategic alliance: Claude embedded into KPMG Digital Gateway, the platform KPMG’s people and clients use to do the actual work. Every one of 276,000 employees across 138 countries gained access. Tax, audit, legal, and advisory. Every practice, every level. No layoffs announced. No headcount reduction attached.

Tim Walsh, Chair and CEO of KPMG US, stated the logic directly: “Our clients depend on us where accuracy, judgment, and knowledge matter most. AI is changing how we deliver — trust and innovation are the core of this new offering and we’re proud to partner with Anthropic to deliver incredible value to our clients and our people.”

That quote is the answer to the question that every other company on this list got wrong.

Now look at what else happened this year.

Oracle eliminated 30,000 positions, notified by email. Meta cut 8,000 after recording some of the strongest profits in the company’s history. Intuit eliminated 3,000 while revenue grew 17%. ClickUp cut 22% of its workforce and offered the remaining employees million-dollar salaries.

Every one of these announcements came with a version of the same explanation: AI had changed the productivity calculus. One engineer with AI tools produces the output of several without. The headcount was adjusted accordingly.

That explanation is not credible.

These are not organizations that lacked the resources to do what KPMG did. Meta’s cuts came after record profits. Intuit’s came while revenue was growing at 17%. Oracle, Meta, Intuit, and ClickUp are not businesses in distress forced into painful triage. The money was there. The talent was there. The technology was the same technology KPMG just put in front of 276,000 people. What was missing was not a resource. It was a vision for how to deploy one.

AI was the explanation. Leadership failure was the cause.

The difference between augmentation and elimination is not a technology decision. It is a leadership decision that reveals what an organization believes its people actually produce. KPMG’s answer is in the Walsh quote: accuracy, judgment, and knowledge. Those are not tasks. AI does not have them. You can give better tools to the people who do, or you can replace the people and discover that the tools produce nothing without the judgment behind them.

The organizations that chose elimination had a different implicit answer to the same question, or they had no answer at all, and reached for headcount reduction because it was the move they knew how to make. Management systems that had never built the capacity to move at the speed AI now makes possible. Organizations that, given the same tools and resources, could not arrive at the same vision. AI became the public explanation for decisions that had different underlying causes.

This is what LeadershipOS™ identifies as the critical distinction in any transition. The organizations that have built a management operating system, where judgment is distributed, context is structural, and every person has clarity about what they produce and why, recognize augmentation immediately. The tool fits into the system. The system absorbs the capability and amplifies it.

The organizations that have not built that system see the same tool and experience it differently. The individual information nodes that have been carrying everything become visible as costs. The absence of structural clarity makes it difficult to see what AI would multiply. So they cut what they can measure, call it an AI transition, and let the workers find out by email.

KPMG did not give AI to 276,000 people because they negotiated a better technology contract. They gave it to 276,000 people because they understood what those 276,000 people produce, and they built a plan to make every one of them more capable at producing it. That is not an AI strategy. It is a leadership strategy that happens to use AI.

The question is not whether to introduce AI to your team. It is what you believe your team actually produces. If the answer is tasks, AI looks like a replacement. If the answer is judgment, AI looks like a multiplier. The operating system you have built gives you the answer before the question arrives.

The LeadershipOS™ Scorecard measures whether your operating system is built to amplify judgment or execute tasks, and what that means for your team’s position in the AI transition. Take it here: https://theleadershiposbook.com/scorecard

Sources: KPMG and Anthropic press release, May 19, 2026 (https://kpmg.com/xx/en/media/press-releases/2026/05/kpmg-and-anthropic-sign-global-alliance-and-launch-digital-gateway-powered-by-claude.html); Anthropic announcement (https://www.anthropic.com/news/anthropic-kpmg); Oracle, Meta, Intuit, and ClickUp layoff figures from public reporting.

---

I write about structural leadership for technical leaders in high-stakes operating environments. If you want to see where your system is load-bearing on you personally, the LeadershipOS™ Scorecard maps it: https://theleadershiposbook.com/scorecard