Why AI Puts the Traditional CPO Role Under Pressure

While commuting last week, I listened to a podcast discussion about which roles are being reshaped first by GenAI and AI-assisted software development.

The obvious answers came up quickly: developers, designers, analysts, product managers. That part was familiar.

The more interesting point was higher up the org chart.

The role that kept sticking with me afterwards was not the engineer, or even the PM. It was the CPO.

For years, many organizations have treated the Chief Product Officer and Chief Technology Officer as parallel but separate authorities. One owns vision, roadmap, customer value, prioritization, and commercial fit. The other owns architecture, engineering capability, delivery quality, and operational reliability.

That split made sense when product definition and product construction were more clearly separated activities.

It makes less sense when AI starts collapsing the distance between idea, prototype, validation, and production.

The line between deciding and building is getting thinner

AI-assisted development is not just making engineers faster. It is changing who can participate meaningfully in the act of building.

Product managers can now test flows, generate drafts, explore edge cases, and pressure-test requirements with far less handoff overhead. Engineers are increasingly expected to think beyond implementation and act more like product builders: people who understand business context, customer trade-offs, and commercial consequences.

Once that happens, the old boundary starts to wobble.

If engineers are expected to move upstream into product thinking, then product leadership also has to move downstream into delivery reality.

You cannot ask the organization to become more integrated while keeping the executive model fragmented.

Why the traditional CPO model is under pressure

To be clear, I am not arguing that product leadership disappears. Far from it.

I am arguing that the traditional interpretation of the CPO role is now under pressure.

The old model often rewarded leaders who were strongest at:

• market framing,
• prioritization,
• roadmap storytelling,
• feature strategy,
• stakeholder alignment.

Those things still matter. But they are no longer sufficient on their own.

In an AI-shaped product organization, the leader at the top needs to understand not only:

• what should be built,
• why it matters,
• who it is for,

but also:

• how it is likely to be built,
• where delivery friction will emerge,
• which parts of the system create leverage,
• how AI changes cost, speed, and quality economics.

That is a different job from classic product stewardship.

It is much closer to product creation leadership than product portfolio management.

The real shift: from product ownership chains to product-building systems

Historically, many companies built isolated chains of ownership:

• strategy writes the ambition,
• product writes the requirements,
• engineering builds the thing,
• sales packages the story,
• operations absorbs the consequences.

Each function can perform well locally while the system still underperforms globally.

AI exposes that weakness faster, because it accelerates the mechanics of delivery while also making coordination gaps more obvious.

If the organization can move from concept to prototype in days instead of months, then poor decisions, vague ownership, and artificial separations become visible much earlier.

The winning organizations will not be the ones with the most AI tools. They will be the ones that build a tighter loop between:

• customer understanding,
• product judgment,
• technical feasibility,
• speed of execution,
• and commercial packaging.

That loop is hard to run when your leadership structure still assumes product and technology are separate factories.

If developers are becoming product builders, executives must evolve too

This is the contradiction many companies are now walking into.

They want engineers who:

• understand customer outcomes,
• think commercially,
• use AI to explore solutions faster,
• contribute to prioritization,
• challenge weak requirements.

At the same time, they still tolerate executive structures where product leadership can remain distant from how products are actually made.

That will not hold.

If the engineering organization is becoming more product-literate, then product leadership has to become more delivery-literate, more systems-literate, and more technically credible.

Not because every CPO must become a CTO.

But because the top product role now needs to understand the economics of product creation, not just the semantics of product strategy.

The industrial lesson: markets were often won by people who understood making and selling

This is not a new pattern. We have seen it before in earlier periods of industrial change.

Josiah Wedgwood: craft, process, and brand in one model

Wedgwood did not succeed simply because he made pottery. He understood product quality, manufacturing consistency, branding, market segmentation, and distribution as one integrated system.

He was not operating with a modern org chart, but the lesson is modern: the leader who understands only the object, or only the market, is weaker than the leader who understands both the making and the selling.

Henry Ford: product design and production design were inseparable

Ford is often remembered for scale and the assembly line, but the deeper lesson is that product and production were strategically linked.

He did not treat manufacturing as a downstream service that merely executed a product vision. The method of making the car was part of the product strategy itself. Cost, accessibility, reliability, and reach all flowed from that integration.

William Lever: packaging, manufacturing, and commercial storytelling together

Lever did not build market leadership in soap by treating production and go-to-market as unrelated disciplines. Product formulation, branding, packaging, and distribution were tightly connected.

Again, different era, same principle: integrated understanding beats functional isolation.

Why SMEs often understand this instinctively

Many SMEs do not have the luxury of artificial boundaries.

Founders in smaller firms often have to understand:

• what customers will pay for,
• what can actually be delivered,
• where cost sits,
• how the offer should be positioned,
• and how the team will execute.

That does not make every SME better run than a large enterprise. It does explain why smaller firms sometimes move with more coherence.

They are forced into end-to-end thinking.

Large organizations often lose that coherence by specializing too early and coordinating too slowly.

AI is punishing that pattern.

What replaces the old CPO model?

Probably not a simple job-title swap.

I do not think every company suddenly needs to abolish the CPO title or merge it formally with the CTO role. Titles are the least interesting part of this discussion.

The real question is this:

Who is accountable for the full system of product creation?

That includes:

• market understanding,
• proposition design,
• roadmap judgment,
• delivery constraints,
• platform leverage,
• AI-enabled workflow design,
• commercial packaging,
• and adoption outcomes.

Some organizations will answer that by evolving the CPO role. Some will create tighter CPO-CTO pairings. Some will move toward a chief product-and-technology model.

But very few will be well served by keeping the old separation intact while expecting everyone else below it to work in a more fused way.

What I would now expect from a strong CPO

In this environment, a strong product leader should be able to speak credibly about:

• customer value and pricing power,
• delivery system constraints,
• platform reuse and technical leverage,
• AI-assisted product discovery,
• AI-assisted delivery economics,
• where human judgment must remain non-negotiable,
• and how the product is taken to market, not just defined on paper.

That is a more demanding brief.

It is also a more relevant one.

A final provocation

For years, product leadership could afford to be strong on intent and lighter on construction.

AI is changing that.

When product discovery, prototype generation, design iteration, engineering implementation, and operational rollout all start compressing into one faster loop, the executive who only understands one side of that loop becomes a bottleneck.

So yes, I think the CPO is one of the roles most exposed to change in the AI era.

Not because product becomes less important.

Because product becomes too important to be led from only one side.

The organizations that adapt fastest will likely be the ones led by people who understand the art of making, the discipline of building, and the craft of selling as parts of the same system.