Start with the question most leaders get backwards. What is middle management actually for?
For decades, the answer was information. Managers relayed it up, pushed it down, and coordinated across. That was the job.
AI now does much of that relay work. So the layers built to move information have less to move. This is why an AI-native organization tends to run flatter.
But flatter is not the goal. It is the byproduct. The goal is to put decisions where the judgment actually lives.
Flattening is already happening at scale
This is not a theory. It is happening inside the largest companies on earth.
Amazon asked every senior team to raise the ratio of individual contributors to managers by 15%. Its CEO wanted fewer layers and faster decisions. He even named the enemy: the “pre-meeting for the pre-meeting.”
So the direction is clear. AI-native organizations are shedding the layers that slow decisions down. Consider what those layers cost you. Every extra layer adds a handoff, a delay, and a diluted decision. So removing them is not just cheaper. In other words, the layer tax is real.
But here is the fork, and it matters more than the trend.
Cutting layers is not the same as redesigning the org
There are two very different ways to flatten. One is reorganization. The other is redesign.
Reorganization cuts layers to save cost. It moves boxes on a chart and calls it change. The decisions still route the same way, just with fewer people. So you get a leaner chart and faster chaos.
Redesign is different. It changes who actually holds the decision. It pushes authority to the people closest to the work. Then the structure follows the decision, not the other way around. This is the structural question. It sits beneath the change method I covered in *Transformation Management is the New Change Management*.
Bayer shows the redesign version at full scale. It cut its layers from about 13 to 6. Then it moved decision-making into roughly 2,000 small, self-directed teams. And it redefined managers as coaches, not controllers.
One Bayer leader framed the whole shift in a single line. Design the organization “around what it means to be human,” not despite it.
The results followed, too. Bayer reports faster decisions, and some teams doubled revenue in a single 90-day cycle. So the flatter structure did more than trim cost. It moved the company closer to its customers.
The middle is where AI transformation gets built
Here is a mistake that flattening can hide. Do not confuse the relay layers with the people who drive change.
AI transformation is not like other technical rollouts. It starts at the top. The C-suite has to buy in fully, because the strategy and the decisions begin there.
But the top does not implement it. The middle does. Managers and team leaders are the evangelists and the doers. They turn strategy into daily practice, and they carry the rest of the organization with them.
So the middle is not dead weight. Rather, it is the engine of the change. Shed the bureaucratic relay, yes. But keep and elevate the leaders who actually build.
That means one thing above all. You upskill both layers. The C-suite has to be ready to decide. The middle has to be ready to deliver. Skip either one, and the transformation stalls.
Flattening without redesign costs more than it saves
Here is the part leaders underestimate. The benefits of an AI-native organization are not automatic. They show up only with intention, a clear strategy, and targeted upskilling for the leaders who remain.
Skip that work, and flattening backfires. We have watched it happen across the market this year.
The pattern is now common enough to have a name: the AI boomerang. Companies cut roles for AI, discover the tool cannot hold the judgment, and quietly rehire.
And the reversal is expensive. Careerminds found that one in three companies spent more on rehiring than they saved. Meanwhile, Forrester found that 55% of employers now regret their AI-driven layoffs.
The named cases make it concrete. Ford rehired hundreds of veteran engineers after automated systems missed quality problems. Commonwealth Bank reversed its customer-service cuts when call volumes actually rose.
So the lesson is blunt. Flatten without redesign, and you pay twice. First you pay for the AI. Then you pay again to rebuild what you cut.
Teach leaders to think differently
Here is the part no reorganization touches. A flatter org asks leaders to lead in a new way.
The old leader owned the decision. The new leader builds the conditions for good decisions. So the job shifts from control to design, and from answers to judgment.
That shift is hard, and it is also learnable. But it does not happen by accident. Instead, you have to teach it deliberately.
Think about what actually changes for the manager. Yesterday, authority came from the title. Today, it comes from how well you enable the team. Therefore the whole definition of a good leader shifts.
Bring the technology in as an amplifier
An AI-native organization does not choose between technology and people. Rather, it uses one to elevate the other.
So let AI carry the coordination load. Let it move the information the layers used to move. Then people rise to the work only a person can do.
That work is judgment, relationship, and decision quality under uncertainty. It is the human layer. And that layer is where the value now compounds. This is where technology and humanity finally stop competing. Instead, each makes the other stronger. The machine handles scale. Meanwhile, the person handles meaning.
The human layer is the moat
Here is why this is a strategy question, not an HR one. As AI commoditizes the routine work, the durable advantage moves to what it cannot copy.
Your moat is no longer the process. It is the quality of judgment distributed across your people. So an AI-native organization protects that layer on purpose. I made the case for that human premium in *The Price Just Went Up on Humanity*.
You flatten to bring people closer to the decision. You bring in AI to free them for judgment. And you build the moat where it actually holds. Because no one can copy that judgment, it holds under pressure. So the flatter, more human org is also the more defensible one.
What this means for real leaders in plain English
Do not flatten your org to cut cost. Flatten it to move decisions to the people who should own them.
So start with the decisions. Ask who holds each one, and who should hold it. Then redesign the structure around that answer, and bring AI in to carry the load.
An AI-native organization is flatter because it trusts its people more. That is the whole point. Above all, remember the order. Decisions come first, structure comes second, and technology comes third.
Where Alchemy comes in
At inflection points, a flat org either compounds value or creates faster chaos. The difference is whether the redesign was human-first.
Alchemy designs AI-native organizations from the decisions out. We redraw decision rights, teach leaders to lead the new way, and place technology where it amplifies people.
If you’re ready for a conversation: https://calendly.com/brynn-alchemyleadershiplab/20min-discovery-call
References
- Amazon, “Update from CEO Andy Jassy on manager team ratio” (Sep 2024) — https://www.aboutamazon.com/news/company-news/ceo-andy-jassy-latest-update-on-amazon-return-to-office-manager-team-ratio
- Harvard Business Review, “Amazon CEO Andy Jassy on Agility, AI Strategy, and the Changing Role of Managers” (May 2025) — https://hbr.org/podcast/2025/05/amazon-ceo-andy-jassy-on-agility-ai-strategy-and-the-changing-role-of-managers
- CIO Dive, “Amazon cuts 14K jobs in AI-fueled reorganization” (Oct 2025) — https://www.ciodive.com/news/aws-layoffs-cloud-AI-andy-jassy/804046/
- Bayer, “Dynamic Shared Ownership” — https://www.bayer.com/en/strategy/dynamic-shared-ownership
- Harvard Business School, “Radical Transformation at Bayer: Dynamic Shared Ownership” (case) — https://www.hbs.edu/faculty/Pages/item.aspx?num=67228
- IMD, “How Bayer’s dynamic path of shared ownership offers lessons for every leader” (2026) — https://www.imd.org/ibyimd/business-transformation/how-bayers-dynamic-path-of-shared-ownership-offers-lessons-for-every-leader/
- CNBC, “Employers who laid off workers citing AI are already starting to regret it” (Jul 2026) — https://www.cnbc.com/2026/07/01/employers-who-laid-off-workers-for-ai-are-reversing-their-decisions.html
- The Washington Times, “Companies rehire for roles eliminated by AI” (Mar 2026, Careerminds/Forrester) — https://www.washingtontimes.com/news/2026/mar/10/ai-layoff-reversal-companies-rehire-customer-roles-eliminated/
- Alchemy Leadership Lab, “The Price Just Went Up on Humanity” — https://alchemyleadershiplab.com/price-just-went-up-on-humanity/
- Alchemy Leadership Lab, “Transformation Management is the New Change Management” — [confirm slug in WordPress]
Estimated reading time: 7 minutes