Autonomy Is the Next Industrial Revolution

In my last article, I argued that healthcare's biggest technology problem isn't artificial intelligence. It's that hospitals don't have an operating system capable of connecting and orchestrating the enormous number of people, devices, applications, and workflows required to deliver care. We keep adding intelligence to healthcare, but much of it lives inside systems that remain disconnected from one another.

The more I've thought about that problem, the more convinced I've become that it isn't unique to healthcare. We are entering a much broader transition in computing, one that will eventually affect transportation, manufacturing, logistics, energy, and nearly every other industry. Artificial intelligence has given machines an extraordinary ability to interpret information, recognize patterns, reason through problems, and increasingly make decisions. The next phase connects that intelligence to action.

That transition is autonomy, and I believe it represents the beginning of the next industrial revolution.

When intelligence becomes action

For most of computing's history, software has occupied a predictable position between information and people. Systems collect data, organize it, analyze it, and present something back to us. Even as software has become dramatically more sophisticated, the relationship has remained largely unchanged: technology tells us something, and a person decides what happens next.

AI has pushed that boundary further. We now have systems that can interpret complex documents, generate software, understand images, communicate in natural language, and reason across enormous amounts of information. Yet most organizations still have a gap between what a machine knows and what the organization actually does. An AI system might recognize that a patient is ready for discharge, equipment needs maintenance, or inventory needs replenishing, but people still spend enormous amounts of time turning that knowledge into action.

Autonomy begins when that boundary starts to disappear. Instead of simply identifying what should happen next, an intelligent system can initiate the appropriate workflow, identify available resources, coordinate with other systems, monitor execution, adapt when conditions change, and involve a human when judgment is required.

AI gives machines intelligence. Autonomy connects that intelligence to execution.

Intelligence is leaving the screen

We can already see this transition happening around us. Autonomous vehicles are learning to interpret complex physical environments and act within them. Robots move inventory through warehouses and perform increasingly sophisticated tasks in factories. Humanoid robotics is advancing as companies try to build machines that can work in spaces originally designed for people. At the same time, AI agents are beginning to execute digital work across applications without waiting for a person to initiate every step.

These developments are often discussed as separate markets: AI, robotics, autonomous vehicles, industrial automation. I think they are different expressions of the same transition. Intelligence is leaving the screen and becoming connected to how work gets done.

Every major technological era has expanded some dimension of human capability. Mechanization multiplied physical strength. Computing multiplied our ability to store and distribute information. Artificial intelligence is multiplying our ability to process knowledge. Autonomy extends that progression into action.

Healthcare makes the opportunity obvious

Healthcare provides an unusually clear example because hospitals already contain extraordinary amounts of technology. A modern health system may have an electronic health record, clinical communication platforms, connected medical devices, real-time location systems, pharmacy automation, mobile devices, transportation teams, robotics, and hundreds of specialized applications. Yet people still provide much of the connective tissue.

Consider something as ordinary as discharging a patient. The physician may determine that the patient is ready to leave, but that decision triggers an operational process involving medications, transportation, family communication, equipment, follow-up appointments, and room turnover. Nurses, case managers, coordinators, and other staff often have to manually move that process forward across disconnected systems.

An autonomous environment approaches the problem differently. Once the appropriate clinical decision has been made, the system can begin coordinating the operational work around it. Pharmacy becomes part of the timeline. Transportation matches the patient's needs. Family members receive appropriate updates. Follow-up care is coordinated. Environmental services understands when the room will become available. The care team remains involved where clinical judgment is required without having to function as the integration layer connecting every operational system in the hospital.

Over time, physical autonomy becomes part of the same environment. Robots can move medications, supplies, and equipment. Intelligent mobility systems can assist with patient movement. Autonomous transportation could eventually extend the workflow beyond the hospital and into the community.

The point isn't to remove humans from healthcare. It is to remove some of the work surrounding healthcare that prevents humans from focusing on care.

From autonomy to orchestration

Another problem waits on the other side of this transition. As organizations deploy more AI agents, robots, connected devices, autonomous vehicles, and intelligent infrastructure, those systems will need coordination. Without an orchestration layer, we risk recreating today's technology fragmentation with hundreds of systems that are now capable of acting independently.

A hospital of the future could have thousands of intelligent participants operating simultaneously. The hard problem won't be making each component intelligent. It will be determining how those components work together, what authority each system has, when humans become involved, and how accountability is maintained.

That connects directly back to the operating system I wrote about previously. The hospital operating system of the future cannot simply connect applications and information. It will eventually need to coordinate action across people, software, machines, vehicles, devices, and physical infrastructure.

Healthcare will not be alone. Manufacturing, logistics, transportation, energy, and other industries will face the same challenge as intelligence becomes embedded throughout their operations.

The next industrial revolution

Industrial revolutions become significant when technology changes how we organize work. Steam power transformed manufacturing and transportation. Electricity changed factories, cities, and homes. Computing reorganized information. Artificial intelligence is changing how we create and apply knowledge.

Autonomy changes the relationship between intelligence and execution.

Humans will remain essential, particularly in healthcare where judgment, empathy, ethics, and accountability matter deeply. But our relationship with technology will change as intelligent systems increasingly participate in the work rather than simply provide information about it.

For the last 30 years, we have been teaching computers how to think. The next 30 will increasingly be about teaching them how to do.

For healthcare, the opportunity is not replacing the human. It is using autonomy to remove coordination, searching, transporting, documenting, and administrative friction around care so people can spend more time doing the work we actually need humans to do.

And that may make healthcare one of the most important proving grounds for the next industrial revolution.

Next month: The Five Ages of Computing, a framework for understanding the progression from Mechanization and Digitization through Intelligence, Autonomy, and ultimately Orchestration.