3 shifts shaping how we engage with intelligent systems: A pioneering technologist explains

I spend my days bringing robots to life. In my studio, industrial robots built to weld chassis and stack pallets learn to notice people, follow a gesture, even chase their own curiosity. Underneath are layers of sensing, motion planning and control, but none of it matters until a person steps into the room. A robot only comes to life when someone is there to meet it. The illusion of sentience needs a human in the loop.

For the 17th Annual Meeting of the New Champions, I created Robots as Mirrors, an installation in which an industrial robot responds to human presence through depth sensing and body language. There are no barriers between visitor and robot. Approach, and it engages; reach out to touch it and it retreats. The encounter feels less like operating a robot and more like meeting an unfamiliar animal.

A decade of building these encounters has convinced me we are asking the wrong question about robots. Something stranger is underway. Robotics is entering an era of magical realism: we are animating the inanimate world. Robots that once lived behind safety cages are stepping into our streets, hospitals and homes.

Three shifts characterize this new era, reshaping not just how machines operate, but how we cohabit with them.

For 50 years, we have reached into the digital world through a pane of glass. Screens, keyboards and mice let us poke and prod at a domain that exists somewhere else. Robots collapse that boundary. They are a bridge between the magic of the digital world and the majesty of our physical one. When an intelligent system shares your space, the interface is no longer a device: it is proximity, posture and movement. It is the same non-verbal language we share with the rest of the living world.

This has practical consequences. If robots are to move safely among people, their behaviour must be legible. A person should be able to read what a robot is doing, why it is doing it, and what it will do next – in the same way we read a driver’s intentions at a crosswalk without exchanging a word. Capability without legibility produces confusion and fear, no matter how impressive the specs. Making a robot’s behaviour readable to human bodies is not a soft problem layered on top of the hard engineering. It is the interface design challenge of the next decade.

I have spent years working with simulated consciousness. The simulation itself is not new. What is new is that it now occupies physical space with us. When a robot pauses, follows or withdraws, people instinctively search for intention behind the movement.

And in that search, they reveal themselves. Some visitors approach with tenderness, some with suspicion, some with the immediate urge to dominate. This is not an engineering phenomenon; it is a cultural one. A robot mirrors your movements. But the instinct it taps is far older: when something moves in response to us, we feel an immediate connection.

For the entire history of automation, humans specified tasks and machines executed them. Autonomy breaks that contract. To unlock its full potential, we have to become comfortable ceding control, handing real agency to non-human entities. Our role shifts from scripting every action to designing the conditions of action: the guardrails, the incentives, the edges of the domain a robot is free to operate within.

This way of thinking is second nature to policy-makers. Nobody commands an economy or scripts a city. You shape conditions, set incentives, define boundaries and let autonomous agents act within them, whether those agents are people, firms or markets. That systemic instinct is deeply developed in governance and still nascent in engineering. On this frontier, the expertise gap runs in the opposite direction from the one we usually assume.

These three shifts do not erase the societal disruptions brought on by automation. Robots, the leading edge of physical AI, will displace labour, and that challenge demands coordinated global policy.

But we are writing the rules of cohabitation right now, in code and in policy at the same time. Written well, those rules point robots not at replacing human work but at expanding human experience: new presences to think with, build with and live alongside. The inanimate world is waking up, and what it becomes is still entirely up to us.

The Forum is spotlighting how innovation moves from breakthrough to scale to impact ahead of Summer Davos in China, 23–25 June 2026. Follow the latest.