Dr. Justinas Mišeikis – Physical AI Goes Corporate. Most Boards Aren’t Ready.

Guest article by Dr. Justinas Mišeikis. Based on his keynote at Cairo Forum 2026, Swiss Innovation Park Dübendorf.

Justinas Mišeikis:

I opened my keynote in Dübendorf with three numbers.

67% of senior executives describe Physical AI as game-changing. That figure comes from a Capgemini survey of 1,678 executives. The World Economic Forum has identified 23 high-potential technology convergence combinations across eight domains. Physical AI sits at the centre of most of them.

The third number is the problem. The share of boards that can articulate a Physical AI strategy is close to zero.

That gap is the most dangerous asymmetry in technology right now. The engineers understand what is coming. The boards do not.

What Physical AI actually is

Generative AI was the warm-up act. It lives on screens. It writes, draws, and summarises. Impressive, but contained.

Physical AI gives software a body. It is artificial intelligence that perceives the world, reasons about it, and acts in it. Robots. Autonomous vehicles. Drones. Machines that see, decide, and move.

The difference matters for one simple reason. A chatbot that makes a mistake produces a bad paragraph. A robot that makes a mistake produces a physical consequence. The stakes change. The economics change. The board conversation must change too.

The market numbers tell the story. Most research firms place the Physical AI market at roughly five billion dollars in 2025. The same firms project growth of 32 to 36 percent per year. That puts the market between 50 and 85 billion dollars by the mid-2030s. The exact figure depends on whether you count the hardware. The direction does not.

Four futures, one choice

The World Economic Forum has mapped four scenarios for how Physical AI deployment could look by 2031.

The first is proven deployment. The technology works, trust is earned, and adoption spreads across industries.

The second is tech disillusionment. Demos impress, but real-world reliability disappoints. Investment cools. The wave stalls.

The third is integrated progress. Deployment succeeds, and the benefits spread broadly across economies and regions.

The fourth is divided deployment. The technology works, but a small number of actors control the data, the supply chains, and the platforms. Everyone else becomes a consumer.

Here is the uncomfortable question I asked the room in Dübendorf:

“Which scenario is your company currently on track for?”

Most European corporations are drifting towards the fourth one. Not by decision. By deferral.

The view from inside the boardroom

I spend my working life between AI research labs and corporate strategy rooms. I have built robots, and I have built the business cases around them. From that seat, I see the same three conversations being postponed again and again.

The first deferred conversation is about data. Physical AI systems learn from physical-world data. Sensor streams, factory layouts, movement patterns, failure logs. Most companies sit on years of this data and treat it as exhaust. Whoever structures it first owns a moat. Boards defer this conversation because it sounds technical. It is not. It is an asset allocation decision.

The second deferred conversation is about make versus partner. Very few companies should build Physical AI capabilities from scratch. Almost all of them should be running structured partnerships with the labs and startups that do. Boards defer this because partnerships feel optional. They are not. The best partners are being signed right now, and exclusivity clauses are real.

The third deferred conversation is about people. Physical AI does not replace a workforce overnight. It changes what the workforce does. The companies that retrain early get a decade of advantage. The companies that wait get a crisis. Boards defer this because the payoff sits beyond the next reporting cycle. That is precisely why it belongs at board level and not in a departmental budget.

None of these conversations requires a single robot to be purchased. All of them require a decision to engage.

Three things to do before the next board meeting

I closed the keynote with three actions. They are deliberately small. The goal is movement, not transformation.

First, run a data audit through a Physical AI lens. Ask one question: what physical-world data do we already generate, and who would pay for it or build on it? Most companies are surprised by the answer.

Second, map your exposure. Identify the three processes in your business that involve repetitive physical work, dangerous physical work, or precision physical work. Those are the entry points. You do not need a strategy for everything. You need a position on three things.

Third, assign ownership. Physical AI cannot live in the IT department, because it is not an IT problem. It cannot live in operations alone, because it reshapes the business model. Someone at executive level needs to own the question. Pick the person before your competitor picks theirs.

The window

Every technology wave has a window in which strategy is cheap. Early enough that positions are still open. Late enough that the technology is real. For Physical AI, that window is open now.

It will not stay open long. The scenarios diverge around 2031. The decisions that determine which scenario your company lives in are being made between now and then. Mostly by default. Occasionally by design.

AI is not an IT problem to solve. It is a strategic business decision to make. Physical AI is the clearest example yet.

The engineers are ready. The question is whether the boardroom will be.

Dr. Justinas Mišeikis is an AI commercialisation strategist based in Zürich. He holds a PhD in Robotics and Computer Vision, an Executive MBA, and more than 40 patents. He is an advisor of the Humanoid Robotics Association and Humanoid Robotics World Championship in Zürich and host of the TechDrive Zürich interview series. He speaks across Europe, the Middle East, and Asia on Physical AI and corporate strategy.

To book Dr. Justinas Mišeikis for a talk on Physical AI: 1 (704) 804 1054 – justinas.miseikis@premium-speakers.com