ctaio.dev Ask AI Subscribe free

AI-NATIVE ORG DESIGN

The AI-Native Org Chart

When pure people managers stop working

Chesky's line on TBPN gets quoted a lot. What it actually implies for the org chart, the IC track, and the CAIO's seat is a different conversation. This is that conversation.

The AI-Native Org Chart: Why Pure People Managers Are at Risk

WHAT CHESKY SAID

The TBPN segment, in context

On TBPN on 2026-05-08, Brian Chesky argued that the era of the pure people manager, the leader whose entire job is to manage other people without contributing to the underlying craft, is ending. He paired the claim with a number that travelled further than the framing: a reported figure of roughly 60% of new Airbnb code now being AI-written. He used the phrase "founder mode" as shorthand for the operating style he believes survives the transition, contrasting it with the managerial drift he is steering Airbnb away from.

The news cycle reduced this to a quote. The actual implication is structural. If a meaningful percentage of the cheap coordination work that middle management used to do is now done by AI, the span-of-control math underneath every org chart changes. Companies that do not redesign around that change end up paying for a layer of management whose work has been quietly automated, while overloading the seniors who used to rely on that layer.

THE THESIS

Pure people managers are the most exposed role

The pure people manager is a role with no IC craft underneath. They run meetings, write performance reviews, forward emails, and translate between layers. Each of these is now substantially augmented by AI. The role does not disappear. It compresses, and the compression hits the layer above first.

The hybrid manager, the one who still ships work alongside managing a team, gets stronger. The IC who used to be capped by a thin coordination layer can now operate at a larger surface. The CAIO\'s job is to design the org so the strong roles get bigger and the weak roles get reabsorbed, rather than letting the change happen through attrition and burnout.

LAYER BY LAYER

Where the pressure lands, by layer

AI compression is not uniform across the org chart. It hits coordination-heavy layers hardest, judgement-heavy layers least. This is the practical read for a CAIO designing the next two years of headcount.

CEO and executive team

Low to medium pressure

What changes: Span widens. Direct AI tooling means the CEO can read deeper into the org without a chief of staff layer.

The risk: Becomes the bottleneck if they refuse to delegate decisions AI now makes faster.

C-suite functional leaders

Medium pressure

What changes: Each function (CFO, CMO, COO) gains an embedded AI partner. The CAIO is the connective tissue.

The risk: Vendor capture. Each function picks a different AI stack and the integration debt compounds.

VPs and senior directors

Medium pressure

What changes: Shift from coordinating people to coordinating AI plus people. Reading dashboards is now a generated summary, not a meeting.

The risk: Identity threat. The role that was about running large teams is now about running smaller teams better.

Managers of managers (directors)

High pressure

What changes: Span widens by 30-60% under good AI tooling. Some directors absorb adjacent areas. Others find themselves redundant.

The risk: This is the layer Chesky is pointing at. Pure people managers without an IC craft are the most exposed.

First-line managers

High pressure

What changes: Team size grows from 6-8 to 10-12 under good AI tooling. Coaching becomes the main job, not coordination.

The risk: Burnout. The administrative load fell, but the emotional and judgement load rose.

Individual contributors

Medium, mostly upward pressure

What changes: Senior ICs become more valuable. Junior IC pipeline shrinks because the cheap-coordination work is gone.

The risk: The IC track has to be elevated structurally, or you create a generational gap with no one in the middle.

WHAT THIS MEANS FOR THE CAIO

Three structural moves for the CAIO

The first move is to defend an elevated IC track. If pure people managers are under pressure, the senior IC has to become a real career destination with comp, scope, and authority that match a director. Most companies do not have this and the gap will hurt them within two cycles. The CAIO needs to push HR and finance to fund it before the senior ICs leave.

The second move is to redesign span of control deliberately rather than reactively. The default reaction is to keep the old structure and run it harder. The better move is to model what each layer should look like after AI tooling at full coverage, then sequence the change through hiring and attrition. Targeting roughly 30-60% wider spans for first-line management, plus a real IC ladder, is a defensible starting point for most software-heavy companies.

The third move is to claim the org-design seat. AI-driven org change cannot live inside HR alone, because HR does not own the tooling decisions that drive the compression. It cannot live inside engineering alone, because engineering does not own people policy. The CAIO is the natural owner because the role already spans both. If the CAIO does not take this seat in 2026, somebody less qualified will fill the vacuum and the redesign will be done badly.

Frequently Asked Questions

Will AI take over middle management?
Not exactly. AI will absorb the parts of middle management that were always automation in disguise: status reports, scheduling, simple performance summaries, routine approvals. What survives is the part of middle management that is actually leadership: judgement under ambiguity, coaching individual contributors through hard problems, defending the team's priorities across the org. The pure people-manager job, where someone manages a team but does not contribute to the work, is the role under pressure. The hybrid manager who still ships gets stronger.
Is middle management necessary in AI-native companies?
Some of it. The span-of-control math changes when AI removes coordination overhead. A senior individual contributor with good AI tooling can carry the equivalent of two reports they used to need a manager to coordinate. Companies that flatten without rebuilding the IC track end up with overworked seniors and no career path. Companies that keep the old org chart end up paying for middle managers who are reduced to forwarding emails. The right answer is fewer managers, stronger IC track, and the CAIO sitting where business and engineering used to fight.
What is the 10-20-70 rule for AI?
The 10-20-70 rule, popularised in BCG and MIT digital-transformation writing and now circulating in 2026 AI-strategy discourse, says roughly 10% of AI value comes from the model, 20% comes from data and infrastructure, and 70% comes from organizational change. It is a rough heuristic, not a measurement. The point is that buying the best model and putting it in front of the same org chart produces almost no compound return. The CAIO who internalizes this number reorganizes the company before they renegotiate the vendor contract.
What is a "manager of managers" and is the role going away?
A manager of managers is a director-level role: someone whose direct reports are themselves managers. It is the layer most exposed to AI compression. When each first-line manager can run a larger team because AI handles coordination, you need fewer managers and therefore fewer managers of managers. The director role does not disappear, but it shifts toward functional leadership and strategy, away from pure span management.
How does the CAIO fit into a flatter org?
In a flatter post-AI org, the CAIO sits as a peer to the CTO and reports to the CEO. The function is small but structurally central: governance, AI talent strategy, and the integration layer between business units and engineering. The CAIO is not a head of a 200-person AI team. The CAIO is the person whose decisions shape how the whole company adopts AI. The team is more often 5 to 20 senior people, plus deep partnerships with every function. See the six CAIO archetypes for which shape fits which org.
·
Thomas Prommer
Thomas Prommer Technology Executive — CTO/CIO/CTAIO

These salary reports are built on firsthand hiring experience across 20+ years of engineering leadership (adidas, $9B platform, 500+ engineers) and a proprietary network of 200+ executive recruiters and headhunters who share placement data with us directly. As a top-1% expert on institutional investor networks, I've conducted 200+ technical due diligence consultations for PE/VC firms including Blackstone, Bain Capital, and Berenberg — work that requires current, accurate compensation benchmarks across every seniority level. Our team cross-references recruiter data with BLS statistics, job board salary disclosures, and executive compensation surveys to produce ranges you can actually negotiate with.

Redesigning the org? Start with the audit.

The readiness audit maps the current span-of-control footprint, identifies the roles most exposed to AI compression, and sequences the redesign.